Object analysis method, device, electronic device and computer-readable storage medium

Abstract

The present application discloses an object analysis method, device, electronic device and computer-readable storage medium. The embodiment of the present application obtains multiple target object groups to be analyzed, and obtains distribution indication parameters of the target object groups on at least one portrait label; based on the distribution indication parameters of the target object groups on the portrait label, the distribution degree of the target object groups on the portrait label is determined; when there are at least two target object groups distributed on the same target label, based on the distribution indication parameters of the at least two target object groups on the target label, the distribution vectors corresponding to the at least two target object groups are determined; based on the distribution vectors corresponding to the at least two target object groups, the distribution difference information of the at least two target object groups on the target label is obtained. The embodiment of the present application can improve the operation effect.

Description

Technical Field

[0001] The present application relates to the technical field of object analysis, and in particular to an object analysis method, device, electronic device and computer-readable storage medium. Background Art

[0002] Under the tide of the Internet, Internet applications are used in all aspects of people's lives. Currently, people can be better served through Internet operations. However, due to the large number of people, the total number of user groups of Internet companies of a certain scale is generally in the tens of millions or even hundreds of millions. Based on operational efficiency considerations, the existing Internet operating methods cannot take corresponding operational actions for user personalization, resulting in poor operational results. Summary of the invention

[0003] The embodiments of the present application provide an object analysis method, device, electronic device and computer-readable storage medium, which can improve operational results.

[0004] In a first aspect, an embodiment of the present application provides an object analysis method, the method comprising:

[0005] Acquire multiple target object groups to be analyzed, and acquire distribution indication parameters of the target object groups on at least one portrait tag, wherein the distribution indication parameters are used to indicate the distribution quantity of the target object groups on at least one tag value to which the portrait tag belongs;

[0006] Determining the distribution degree of the target object group on the portrait label based on the distribution indication parameter of the target object group on the portrait label;

[0007] When there are at least two target object groups distributed on the same target label, determining distribution vectors corresponding to the at least two target object groups respectively based on distribution indication parameters of the at least two target object groups on the target label;

[0008] Based on the distribution vectors respectively corresponding to the at least two target object groups, distribution difference information of the at least two target object groups on the target labels is obtained.

[0009] In a second aspect, an embodiment of the present application further provides an object analysis device, the device comprising:

[0010] An acquisition module, used to acquire multiple target object groups to be analyzed, and to acquire distribution indication parameters of the target object groups on at least one portrait tag, wherein the distribution indication parameters are used to indicate the distribution quantity of the target object groups on at least one tag value to which the portrait tag belongs;

[0011] A distribution determination module, configured to determine the distribution degree of the target object group on the portrait tag based on the distribution indication parameter of the target object group on the portrait tag;

[0012] A vector determination module, used for determining distribution vectors corresponding to the at least two target object groups respectively based on distribution indication parameters of the at least two target object groups on the target label when there are at least two target object groups distributed on the same target label;

[0013] A situation determination module is used to obtain distribution difference information of the at least two target object groups on the target tags based on the distribution vectors respectively corresponding to the at least two target object groups.

[0014] In a third aspect, an embodiment of the present application further provides an electronic device, comprising a memory storing a plurality of instructions; the processor loads instructions from the memory to execute the steps of any object analysis method provided in the embodiment of the present application.

[0015] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a plurality of instructions, wherein the instructions are suitable for loading by a processor to execute the steps in any one of the object analysis methods provided in the embodiments of the present application.

[0016] In an embodiment of the present application, by obtaining multiple target object groups to be analyzed, and obtaining distribution indication parameters of the above-mentioned target object groups on at least one portrait label, the above-mentioned distribution indication parameters are used to indicate the distribution quantity of the target object group on at least one label value belonging to the above-mentioned portrait label; based on the distribution indication parameters of the above-mentioned target object group on the above-mentioned portrait label, the distribution degree of the above-mentioned target object group on the above-mentioned portrait label is determined; when there are at least two target object groups distributed on the same target label, based on the distribution indication parameters of the above-mentioned at least two target object groups on the above-mentioned target label, the distribution vectors corresponding to the above-mentioned at least two target object groups are determined; based on the distribution vectors corresponding to the above-mentioned at least two target object groups, the distribution difference information of the above-mentioned at least two target object groups on the above-mentioned target label is obtained, thereby improving the operation effect through the distribution degree of a single target object group on the portrait label, and the distribution difference information between multiple target object groups on the target label. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0018] Figure 1 This is a flow chart of an embodiment of the object analysis method provided in the embodiment of the present application;

[0019] Figure 2 This is a schematic diagram of distribution difference information of two object groups provided in an embodiment of the present application;

[0020] Figure 3 This is another schematic diagram of distribution difference information between two object groups provided in an embodiment of the present application;

[0021] Figure 4 is a schematic diagram of the structure of an object analysis device provided in an embodiment of the present application;

[0022] Figure 5 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0024] Before explaining the embodiments of the present application in detail, some terms involved in the embodiments of the present application are first explained.

[0025] Among them, in the description of the embodiments of the present application, the terms "first", "second", etc. may be used in this article to describe various concepts, but unless otherwise specified, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another concept. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] The embodiment of the present application provides an object analysis method, device, electronic device and computer-readable storage medium. Specifically, the object analysis method of the embodiment of the present application can be performed by an electronic device, wherein the electronic device can be a terminal or a server. The terminal can be a terminal device such as a smart phone, a tablet computer, a laptop computer, a touch screen, a game console, a personal computer (PC, Personal Computer), a personal digital assistant (Personal Digital Assistant, PDA), etc. The terminal can also include a client, which can be a game application client, a browser client carrying a game program, or an instant messaging client. The server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and basic cloud computing services such as big data and artificial intelligence platforms.

[0027] For example, the electronic device is explained by taking a terminal as an example. The terminal can obtain multiple target object groups to be analyzed, and obtain distribution indication parameters of the above-mentioned target object groups on at least one portrait label, wherein the above-mentioned distribution indication parameters are used to indicate the distribution quantity of the target object group on at least one label value belonging to the above-mentioned portrait label; based on the distribution indication parameters of the above-mentioned target object group on the above-mentioned portrait label, determine the distribution degree of the above-mentioned target object group on the above-mentioned portrait label; when there are at least two target object groups distributed on the same target label, based on the distribution indication parameters of the above-mentioned at least two target object groups on the above-mentioned target label, determine the distribution vectors corresponding to the above-mentioned at least two target object groups respectively; based on the distribution vectors corresponding to the above-mentioned at least two target object groups respectively, obtain distribution difference information of the above-mentioned at least two target object groups on the above-mentioned target label.

[0028] Based on the above problems, the embodiments of the present application provide an object analysis method, device, electronic device and computer-readable storage medium, which can improve operational results.

[0029] The following is a detailed description of the embodiments in conjunction with the accompanying drawings. It should be noted that the description order of the following embodiments is not intended to limit the preferred order of the embodiments. Although the logical order is shown in the flow chart, in some cases, the steps shown or described may be performed in an order different from that shown in the accompanying drawings.

[0030] In this embodiment, a terminal is taken as an example for description. This embodiment provides an object analysis method, such as Figure 1 As shown, the specific process of the object analysis method can be as follows:

[0031] 101. Obtain multiple target object groups to be analyzed, and obtain distribution indication parameters of the target object groups on at least one portrait tag, wherein the distribution indication parameters are used to indicate the distribution quantity of the target object groups on at least one tag value to which the portrait tag belongs.

[0032] Among them, the above-mentioned target object group is an object group that needs to be analyzed in order to take corresponding operational measures based on the analysis results. The target object group is an object group divided based on certain attributes, that is, the target object groups have certain commonalities, so that the terminal can perform analysis based on the commonalities of the target object group to meet the personalized needs of users and make investments more reasonably, thereby improving operational results at a lower cost.

[0033] Furthermore, in order to analyze the commonalities of the target object group, it is necessary to obtain relevant information of the target object group. In this regard, in the present embodiment, the terminal refers to the portrait tag to analyze the target object group. Since there is at least one label value corresponding to different portrait tags, in the present embodiment, the terminal needs to obtain a distribution indication parameter of the target object group on at least one portrait tag. The distribution indication parameter indicates the distribution quantity of the target object group on at least one label value to which the portrait tag belongs, that is, the number of object groups in the target object group that meet a certain label value, so as to analyze the target object group through the distribution indication parameter.

[0034] Exemplarily, the above-mentioned portrait label is set to be a gender label, then the corresponding label value can be male, female and unknown, and the target object group has 50 people, then there can be 40 people when the label value is male, and there can be 10 people when the label value is unknown.

[0035] In some embodiments, the above-mentioned acquisition of multiple target object groups to be analyzed may include: the terminal acquires basic information and behavior information of multiple users through a preset portrait system, and performs information processing based on the basic information and behavior information of the multiple users, assigns at least one portrait tag to each of the multiple users, and the tag value corresponding to the user on the portrait tag, so as to analyze the user through the user portrait system. Then, multiple object circle selection tags and circle selection tag values ​​corresponding to the object circle selection tags are acquired, so as to divide the multiple users into groups based on the multiple object circle selection tags, the circle selection tag values ​​corresponding to the object circle selection tags, and the portrait tags corresponding to the multiple users, and obtain multiple target object groups to be analyzed.

[0036] It can be understood that the number of the circled label values ​​is less than or equal to the number of all label values ​​of the object circled label to which the circled label value belongs. For example, if the group circled label is a gender label, then the circled label value can be female.

[0037] Among them, the above-mentioned object circle selection label and the circle selection label value corresponding to the object circle selection label can be pre-set in the terminal, or can be issued by relevant staff in real time on the display page of the terminal or other terminals associated with the terminal, or can be generated in real time based on the business needs currently input into the terminal. The specific setting can be made according to the needs and is not limited here.

[0038] It is understandable that the terminal can select a target object group through one object selection label, or can select a target object group through two or more object selection labels, and different target object groups may have the same portrait labels. Generally, in order to better analyze the target object group, the number of object selection labels used to select a target object group should be less than or equal to a preset label threshold. The label threshold can be set according to needs and is not limited here.

[0039] In some embodiments, the terminal can configure various types of portrait tags for the portrait system based on relevant technologies and related businesses, and can also form a tag library of various types of portrait tags to meet various business scenarios.

[0040] For example, the above-mentioned multiple types of portrait tags can be: basic portrait tags that include log collection or external purchase and fusion, statistical portrait tags based on historical behavior, mining portrait tags based on historical behavior, and predictive portrait tags, etc.

[0041] In addition, the terminal may also classify the portrait tags into different types according to the types of values ​​that the portrait tags can take, such as enumeration type, continuous value type and date type.

[0042] Among them, the values ​​of the above enumerated type portrait tags are a finite set, and the logical relationship between the portrait tags and the tag values ​​is equal to or not equal to.

[0043] Among them, the value range of the above-mentioned continuous value type portrait tags is infinite in most cases, but it is limited in a few cases. For example, when facing the portrait tag of the number of active days in the past X days, the label value is 0-X. Generally, the portrait tags of continuous value types are positive numbers, which can be integers or decimals. The specific accuracy can be determined according to the actual situation. Although the values ​​of continuous value type portrait tags are infinite in most cases, since the total number of users is a finite set after all, and the values ​​of these users on a certain portrait tag are actually limited, the difference is only based on the cardinality of the value.

[0044] Specifically, for portrait tags of continuous value type, there can be several processing methods in the actual business system. For example, the values ​​can be segmented. For example, for the portrait tag of the number of views, the values ​​can be divided into 5 segments: 0-10, 10-30, 30-50, 50-100, >100, etc., and then it can be converted into an enumeration value type tag. It is also possible to determine a range that can cover most business scenarios, and the remaining values ​​can be considered as abnormal values ​​and discarded. For example, 0-100 is determined to be normal, and the remaining values ​​can be considered as abnormal values ​​and discarded. Therefore, for portrait tags of continuous value type, there are six logical relationships between portrait tags and tag values: greater than, greater than or equal to, less than, less than or equal to, equal to, and not equal to.

[0045] Among them, the value of the above-mentioned date-type portrait tag is a date, which can be the current date, a date in the past, or a date in the future. The value range of the date-type portrait tag and the processing in the actual business system are basically the same as the processing of the continuous value type portrait tag, and a finite set of portrait tag values ​​can be determined according to the actual business scenario. There are six types of logical relationships between portrait tags and tag values: greater than, greater than or equal to, less than, less than or equal to, equal to, and not equal to.

[0046] In some embodiments, the subject of the above-mentioned portrait system can be any entity, such as a general user portrait, note portrait or product portrait, etc., or a subject in a specific field, such as a house portrait, buyer portrait, seller portrait or tenant portrait in the real estate field, such as a passenger portrait, driver portrait or vehicle source portrait in the travel field, such as a rider portrait, merchant portrait, food portrait in the takeaway field. In this embodiment, in order to facilitate the analysis of users in multiple fields, the terminal can use user portraits as the subject.

[0047] Among them, the basic information of the above-mentioned users may be information such as gender, age, occupation, etc., and the user's behavior information may be the behavior written by the user in a certain business scenario, such as browsing behavior, transaction behavior, consulting behavior, etc. In the above-mentioned portrait system, each user has a globally unique identifier to facilitate the identification of the user. The user's globally unique identifier can be used to retrieve all portrait tags belonging to the user to obtain relevant information of the user.

[0048] Specifically, the above-mentioned portrait system can extract the dimension information of at least one characteristic dimension based on the basic information and behavior information of each user, and label the user based on the dimension information of at least one characteristic dimension, that is, assign at least one portrait label to the user. Moreover, through continuous data accumulation and label development, a complete user portrait system can be formed for the above-mentioned portrait system, which can include multiple types of portrait labels.

[0049] 102. Determine the distribution degree of the target object group on the portrait tag based on the distribution indication parameter of the target object group on the portrait tag.

[0050] It is understandable that due to the uncertainty between the portrait tag and the label value to which the portrait tag belongs, such as some portrait tags have only one label value, while some portrait tags have multiple label values, the logical relationship between the label and the label value is also uncertain, such as equal to, not equal to, greater than, less than, etc. Therefore, in order to clearly define the distribution degree of the target object group on the portrait label, and for a specific group, the value distribution of this group on a certain label is determined, that is, the number of people taking the value is determined. Therefore, in this embodiment, the terminal can determine the distribution degree of the target object group on the portrait label by the number of object groups with different label values ​​of the portrait label, that is, the distribution of the target object group on the portrait label is obtained, and the characteristics of the target object group are obtained.

[0051] In this embodiment, by determining the distribution degree of the object group on the portrait label, the distribution of the target object group on different portrait labels is quantified to obtain the personalized needs of the object group, and corresponding operational actions are taken based on the personalized needs of the object group, thereby improving the user experience and realizing refined operations.

[0052] Among them, the results included in the above distribution degree can be set according to needs, for example, the distribution degree of the target object group on the portrait label is expressed by a numerical value in an interval, such as 1 to 10. When it is 1, it means that the target object group is more dispersed on the portrait label, and when it is 10, it means that the target object group is more concentrated on the portrait label. The more concentrated the distribution degree of the target object group on the portrait label, the more commonalities the target object group has on the portrait label. In addition, the results included in the above distribution degree can also be text, such as concentrated, relatively concentrated, relatively dispersed, dispersed, etc., which can be set according to specific needs and are not limited here.

[0053] Since the purpose of refined operation is to specify and implement corresponding plans for a certain object group, in order to better carry out refined operation, operation can be carried out based on the target object group to which portrait tags with a concentrated distribution degree belong, or, based on the target object group to which portrait tags with a dispersed distribution degree belong, and the target object group can be subdivided based on the portrait tags with a dispersed distribution degree, so as to further narrow the scope of the target object group and carry out more personalized operations to achieve better business results.

[0054] In some embodiments, the target object groups are sorted based on their distribution levels on different portrait labels, with those with more concentrated distribution levels placed in front and those with more dispersed distribution levels placed in the back, so that relevant staff can view the commonalities of the target object groups.

[0055] In some embodiments, the determining the distribution degree of the target object group on the portrait tag based on the distribution indication parameter of the target object group on the portrait tag may include: the terminal may determine the probability of the target object group corresponding to at least one tag value based on the distribution quantity of the target object group on at least one tag value to which the portrait tag belongs. Then, the terminal may use a preset statistical algorithm to determine the distribution degree of the target object group on the portrait tag based on the probability of the target object group corresponding to at least one tag value.

[0056] The above statistical algorithms include but are not limited to mean error, variance, standard deviation or information entropy.

[0057] Specifically, determining the distribution degree of the target object group on the portrait label based on the probability that the target object group corresponds to at least one label value includes: calculating the information entropy of the target object group on the portrait label based on the probability that the target object group corresponds to at least one label value, and determining the information entropy as the distribution degree of the target object group on the portrait label.

[0058] It can be understood that the above information entropy refers to the average amount of information after excluding redundancy in the information, which solves the problem of quantitative measurement of information.

[0059] The calculation formula of the above information entropy can be:

[0060]

[0061] Among them, if a portrait label is set to have n label values, then the value results of the n label values ​​can be U1, U2, ..., Un, then the corresponding probabilities of the n label values ​​are P1, P2, ..., Pn, and the label values ​​are independent of each other.

[0062] Exemplarily, it is assumed that there are 100 people in the target group, and on the portrait label A, 40 people in the target group have label value A1, 10 people have label value A2, and 50 people have label value A3. In addition, on the portrait label B, 20 people in the target group have label value B1, 30 people have label value B2, 20 people have label value B3, and 30 people have label value B4.

[0063] Then, for the image label A, the information entropy of its label distribution is:

[0064] H(U) A =-(0.4×log20.4+0.1×log20.1+0.5×log20.5)

[0065] =-(0.4×log0.4 / log2+0.1×log0.1 / log2+0.5×log0.5 / log2)

[0066] =1.361

[0067] For the image label B, the information entropy of its label distribution is

[0068] H(U) B =-(0.2×log20.2+0.3×log20.3+0.2×log20.2+0.3×log20.3)

[0069] =-(0.2×log0.2 / log2+0.3×log 0.3 / log2+0.2×log0.2 / log 2

[0070] +0.3×log 0.3 / log2)=1.971

[0071] It can be seen from the above that the information entropy of the target object group corresponding to portrait label A is 1.361, while the information entropy of the target object group corresponding to portrait label B is 1.971. Since the smaller the information entropy of the target object group on a portrait label, the more concentrated the distribution of the target object group on this portrait label, the distribution degree of the target object group on portrait label A is more concentrated than the distribution degree of the target object group on portrait label B, and the distribution degree of the target object group on portrait label B is more disordered.

[0072] In some embodiments, the larger the information entropy of the target object group on a portrait tag, the more dispersed the distribution of the target object group on the portrait tag, and the more different the target object group is. Therefore, if the information entropy is greater than or equal to the first preset information threshold, it means that the target object group is more dispersed on the portrait tag to which the information entropy belongs, and the target object group is divided into groups based on the portrait tag to which the information entropy belongs, that is, the target object group is subdivided using the portrait tag to which the information entropy belongs, so as to obtain a more refined object group by adding such portrait tags in the selection process.

[0073] If, however, the smaller the information entropy of the target object group on a portrait label, the more concentrated the distribution of the target object group on this portrait label, the more commonality the target object group has. Therefore, if the information entropy is less than or equal to the second preset information threshold, it means that the target object group is more concentrated on the portrait label to which the information entropy belongs, and the target object group is not divided into groups based on the portrait label to which the information entropy belongs, that is, the portrait label to which the information entropy belongs is not used to subdivide the target object group. Among them, the first preset information threshold is greater than the second preset information threshold.

[0074] For example, it is assumed that there is currently a first target group, in which there are four users with user IDs A, B, C, and D. A, B, C, and D included in the target group are stored in the first target group table, and corresponding portrait tags are set, which are tags 1 to 6 respectively. The specific configuration is shown in the following first portrait table:

[0075]

[0076]

[0077] Among them, the above-mentioned first portrait table is a full portrait table, and the value after the label in the table is the label value. From the above table, we can see the distribution of A, B, C, and D in the first target group on labels 1 to 3. In addition, since the above-mentioned first portrait table is a full portrait table, and labels 4 to 6 are portrait labels distributed to other target groups, not portrait labels distributed to the first target group, they have no values ​​in this example.

[0078] The first step is to extract the corresponding data from the table for subsequent processing, that is, to join the first target group table and the first portrait table, that is, to join the conditions of equal primary key columns to obtain the processed first table, which is the object group portrait table to be analyzed. The first table is as follows:

[0079] Primary Key 1 Tag 1 Tag 2 Tag 3 A Tag 1 value -1 Tag 2 value -1 Tag 3 value -2 B Tag 1 value -2 Tag 2 value -1 Tag 3 value -2 C Tag 1 value -3 Tag 2 value -2 Tag 3 value -2 D Tag 1 value -4 Tag 2 value -3 Tag 3 value -3

[0080] In the second step, each row in the first table is processed, and the portrait label and label value are assembled into a json in the kv format. The processed data has only two columns, and the processed second table is obtained. The second table is as follows:

[0081]

[0082]

[0083] The third step is to perform flatmap processing on the data in the second table to flatten the kv structure in json, that is, to convert one row into multiple rows, and obtain the processed third table, which is as follows:

[0084]

[0085]

[0086] The fourth step is to process the data in the third table and swap the kv values, that is, swap the primary key columns and labels, to obtain the processed fourth table, which is as follows:

[0087] "Label 1":"Label 1 value - 1" A "Label 2":"Label 2 value - 1" A "Label 3":"Label 3 value - 2" A "Label 1":"Label 1 value - 2" B "Label 2":"Label 2 value - 1" B "Label 3":"Label 3 value - 2" B "Label 1":"Label 1 value - 3" C "Label 2":"Label 2 value - 2" C "Label 3":"Label 3 value - 2" C "Label 1":"Label 1 value - 4" D "Label 2":"Label 2 value - 3" D "Label 3":"Label 3 value - 3" D

[0088] The fifth step is to group and count the tag values ​​in the fourth table according to the key (tag), or directly group and count the second column in the third table. The key of the group is the tag name + tag value, and the number of people under each tag value is obtained, that is, the processed fifth table is obtained. The fifth table is as follows:

[0089]

[0090]

[0091] Among them, the middle column of the fifth table can be generated according to the needs.

[0092] Step 6: split the data in the fifth table to obtain the first target table, which is as follows:

[0093] Label Column Label value column count "Label 1" "Label 1 value -1" 1 "Label 1" "Label 1 value -2" 1 "Label 1" "Label 1 value -3" 1 "Label 1" "Label 1 value -4" 1 "Label 2" "Label 2 value -1" 2 "Tag 2" "Label 2 value -2" 1 "Label 2" "Label 2 value -3" 1 "Label 3" "Label 3 value -2" 3 "Tag 3" "Label 3 value -3" 1

[0094] Further, after obtaining the above first target table, the terminal can determine the information entropy of the first target object group on labels 1 to 3 according to the above formula for calculating information entropy, and the obtained result table is as follows:

[0095]

[0096]

[0097] In some embodiments, when conducting insight analysis on a single target object group, in order to better meet the needs, the user may also set characteristic variable information so that the terminal introduces the characteristic variable information in the process of determining the distribution degree of the target object group on the portrait label based on the distribution indication parameters of the target object group on the portrait label, so as to comprehensively consider the characteristic variable information and the distribution indication parameters of the target object group on the portrait label to determine the distribution degree of the target object group on the portrait label.

[0098] The above-mentioned characteristic variable information can be set according to user needs, such as the user's city, gender and other information.

[0099] Exemplarily, based on the above example, it is also possible to set the four users A, B, C, and D currently existing in the first target group table, and further analyze the first target group based on the city, and set corresponding portrait tags, which are tags 7 to 11, respectively, wherein the above cities are set to include cities a, b, and c. The specific configuration is shown in the following second portrait table:

[0100]

[0101]

[0102] Among them, the second portrait table is a full portrait table, and the value after the label in the table is the label value. From the above table, we can see the distribution of A, B, C, and D in the first target group on labels 9 to 11 based on the city condition. In addition, since the second portrait table is a full portrait table, and labels 7 to 8 are the portrait labels distributed by other target groups, and the portrait labels distributed by non-first target groups based on city conditions, they have no values ​​in this example.

[0103] The first step is to extract the corresponding data from the table for subsequent processing, that is, to join the first target group table and the second portrait table, that is, to join the conditions of equal primary key columns to obtain the processed first table, which is the object group portrait table to be analyzed. The sixth table is as follows:

[0104] Primary Key 1 - User ID Primary Key 2 - City Tag 9 Tag 10 Tag 11 A a Tag 9 value -1 Label 10 value -1 Tag 11 value -1 A b Tag 9 value -3 Tag 10 value -2 Tag 11 value -2 A c Tag 9 value -2 Tag 10 value -3 Tag 11 value -2 B c Tag 9 value -3 Tag 10 value -2 Tag 11 value -1 B b Tag 9 value -1 Tag 10 value -3 Tag 11 value -1 C c Tag 9 value -2 Tag 10 value -1 Tag 11 value -3 D a Tag 9 value -2 Label 10 value -1 Tag 11 value -2 D b Tag 9 value -4 Tag 10 value -3 Tag 11 value -2 D d Tag 9 value -1 Tag 10 value -3 Tag 11 value -1

[0105] In the second step, process each row in the sixth table, assemble the portrait label and label value into a json in the kv format, and the processed data has only 3 columns, the first 2 columns are primary key columns, and the processed seventh table is obtained, which is as follows:

[0106]

[0107]

[0108]

[0109] The third step is to perform flatmap processing on the data in the seventh table to flatten the kv structure in json, that is, to convert one row into multiple rows, and obtain the processed eighth table, which is as follows:

[0110]

[0111]

[0112] The fourth step is to group and count the tag values ​​in the eighth table according to the key (tag + city), or directly group and count the second column in the third table. The key of the group is city + tag name + tag value, and the number of people under each tag value is obtained, that is, the processed ninth table is obtained. The ninth table is as follows:

[0113]

[0114]

[0115] Among them, the middle column of the ninth table can be generated according to the needs.

[0116] The fifth step is to split the data in the ninth table to obtain the second target table, which is as follows:

[0117]

[0118]

[0119] It can be understood that when the city column is selected as the filtering condition, the information entropy of the first target object group on each label is arranged in order from large to small, and the distribution degree of each label on each label value in a specific city / without distinguishing cities can be obtained, that is, the insight analysis of a single object group is completed.

[0120] As can be seen from the above example, the data in the first target table is not divided by city, while the data in the second target table is divided by city. Therefore, if you want to process the first target group conveniently and merge the data in the two tables together, you need to add a city column to the first target table. The result of the fusion is the third target table, which is shown below:

[0121]

[0122]

[0123]

[0124] It is understandable that the first portrait table and the second portrait table in the above two examples can be analyzed and processed separately at the same time, and multiple labels in the same table can also be analyzed at the same time to improve the analysis speed.

[0125] In addition, the above example can be calculated using a preset offline computing engine, such as the spark offline computing engine.

[0126] 103. When there are at least two target object groups distributed on the same target tag, determine distribution vectors corresponding to the at least two target object groups respectively based on distribution indication parameters of the at least two target object groups on the target tag.

[0127] It is understandable that when facing object groups, some object groups have less feedback effects on the same operation activity, while some object groups have poor feedback effects on the same operation activity. In order to locate the reasons for the differences in feedback between the two object groups on the operation activities, in this embodiment, the terminal can call the same portrait label associated with at least two target object groups as the target label, thereby quantifying the distribution of at least two target object groups in the same dimension, that is, based on the distribution indication parameters of the at least two target object groups on the target label, determine the distribution vectors corresponding to the at least two target object groups respectively, and the distribution vector is the quantified distribution of the at least two target object groups in the same dimension.

[0128] For example, by taking one of at least two target object groups as the benchmark object group and the other object groups as the experimental object groups, by quantifying the distribution difference information between the labels, we can gain insight into which labels have a large difference in distribution between the experimental object group and the benchmark object group, that is, gain insight into the characteristics of the object groups, provide guidance for operational activities, and derive the direction and focus of refined operations.

[0129] It can be understood that, in this embodiment, the terminal can imagine all values ​​of the target tag as a vector space, and the distribution of each target object group on each tag value of the target tag forms a vector in the vector space.

[0130] Specifically, the at least two target object groups include a first object group and a second object group. The above-mentioned determination of the distribution vectors corresponding to the at least two target object groups respectively based on the distribution indication parameters of the at least two target object groups on the target label may include: constructing a coordinate system based on at least one target label value to which the target label belongs, wherein each of the target label values ​​serves as a coordinate axis in the coordinate system; determining a first vector corresponding to the first object group in the coordinate system based on the distribution quantity corresponding to the first object group on at least one of the target label values; and determining a second vector corresponding to the second object group in the coordinate system based on the distribution quantity corresponding to the second object group on at least one of the target label values.

[0131] For example, Figure 2 As shown in the figure, assuming that the target tag A has 2 values, the total number of people in the first object group is x1+y1, there are x1 users at value 2 of the target tag A, and y1 users at value 1 of the target tag A. The total number of people in the second object group is x2+y2, there are x2 users at value 2 of the target tag A, and y2 people at value 1 of the target tag A.

[0132] For example, Figure 3 As shown in the figure, assuming that the target tag B has 2 values, the total number of people in the first object group is x1+y1, there are x1 users at value 2 of the target tag B, and there are y1 users at value 1 of the target tag B. The total number of people in the second object group is x2+y2, there are x2 users at value 2 of the target tag B, and there are y2 people at value 1 of the target tag B.

[0133] 104. Obtain distribution difference information of the at least two target object groups on the target label based on the distribution vectors respectively corresponding to the at least two target object groups.

[0134] It can be understood that since the terminal can imagine all values ​​of the target label as a vector space, and the distribution of each target object group on the various label values ​​of the target label forms a vector in the vector space, therefore, in this embodiment, the terminal can determine the difference between the two in the same vector space based on the distribution vectors corresponding to the at least two target object groups respectively.

[0135] For example, the terminal can quantify the distribution difference information of different target object groups on the target label by calculating the angle between the vectors. The smaller the angle between the vectors, the smaller the difference in the distribution of the label between the different target object groups. The larger the angle, the greater the difference in the distribution of the label between the different object groups.

[0136] Specifically, the above-mentioned obtaining distribution difference information of the at least two target object groups on the target label based on the distribution vectors corresponding to the at least two target object groups respectively may include: calculating the angle information between the first vector and the second vector, and determining the distribution difference information of the first object group and the second object group on the target label based on the angle information.

[0137] The angle information may be the angle between two vectors, or a trigonometric function corresponding to the angle between two vectors, such as sin, cos, etc.

[0138] It can be understood that if the above angle information is a sin value, the larger the sin value, the greater the difference; if the above angle information is a cos value, the larger the cos value, the smaller the difference, and the smaller the cos value, the greater the difference.

[0139] For example, based on Figure 2 and Figure 3 From the example presented in , it can be seen that the angle between the distribution of the first object group and the second object group on target label A is greater than the angle between the distribution on target label B. Therefore, it can be analyzed that the two object groups have a large difference in target label A. In the label selection of the refined operation process, target label A may have a higher value.

[0140] Optionally, since there may be multiple at least two target object groups, the terminal can randomly or according to a preset selection rule select one of the at least two target object groups as the benchmark object group, and then use the other target object groups as experimental object groups, thereby comparing the distribution vectors of the at least two experimental object groups with the benchmark object group respectively to determine the differences between different target object groups in the same vector space.

[0141] For example, if it is assumed that there are currently object group A and object group B distributed on target label 1 and target label 2, and the distribution indication parameters for setting the distribution of object group A on target label 1 and target label 2 are as follows:

[0142] Label Column Label value column Count Column "Label 1" "Label 1 value -1" 1 "Label 1" "Label 1 value -2" 1 "Label 1" "Label 1 value -3" 1 "Label 1" "Label 1 value -4" 1 "Label 2" "Label 2 value -1" 2 "Label 2" "Label 2 value -2" 1 "Label 2" "Label 2 value -3" 1

[0143] The distribution indicator parameters for setting the distribution of object group B on target label 1 and target label 2 are as follows:

[0144] Label Column Label value column Count Column "Label 1" "Label 1 value -1" 1 "Label 1" "Label 1 value -2" 2 "Label 1" "Label 1 value -3" 2 "Tag 2" "Label 2 value -1" 2 "Tag 2" "Label 2 value -2" 4 "Tag 2" "Label 2 value -3" 1

[0145] Then the processing results for object group 1 are shown in the following table:

[0146]

[0147] Then the processing results for object group 2 are shown in the following table:

[0148]

[0149]

[0150] Then, the processing results of object group 1 and object group 2 are joined according to the label column, and the results are shown in the following table:

[0151]

[0152] Since each row of data is the distribution of two object groups under the label, imagine the label distribution as a vector space, where the number of label values ​​is the dimension of the vector space. The distribution of label values ​​for each object group is a vector in the space, and the distribution of the number of people on each label value is the component of the vector on each coordinate axis.

[0153] Then, taking label 1 as an example, we take a union of the label values ​​of the two object groups and remove duplicates, and we get four possible values ​​of label 1. Assume there is a 4-dimensional vector space, where the vector distributed by object group 1 is <1,1,1,1>, and the vector distributed by object group 2 is <1,2,2,0>. By calculating the cosine value of the angle between the two vectors, we can compare the distribution difference information of the two object groups, as shown in the following table:

[0154]

[0155] Finally, the final results can be saved in a table, as shown below:

[0156]

[0157] Among them, the larger the cosine angle value is, the smaller the angle between the distribution vectors corresponding to the two object groups is, that is, the closer the vector directions are, that is, the closer the distribution of the two object groups on the target label is, the smaller the difference is, and vice versa.

[0158] It can be seen from the above content that by obtaining multiple target object groups to be analyzed, and obtaining the distribution indication parameters of the above target object groups on at least one portrait label, the above distribution indication parameters are used to indicate the distribution quantity of the target object group on at least one label value belonging to the above portrait label; based on the distribution indication parameters of the above target object group on the above portrait label, the distribution degree of the above target object group on the above portrait label is determined; when there are at least two target object groups distributed on the same target label, based on the distribution indication parameters of the at least two target object groups on the above target label, the distribution vectors corresponding to the at least two target object groups are determined; based on the distribution vectors corresponding to the at least two target object groups, the distribution difference information of the at least two target object groups on the target label is obtained, thereby improving the operation effect through the distribution degree of a single target object group on the portrait label, and the distribution difference information between multiple target object groups on the target label.

[0159] In order to better implement the above method, the embodiment of the present application also provides an object analysis device, which can be specifically integrated into an electronic device, such as a computer device, which can be a terminal, a server, or other device.

[0160] Among them, the terminal can be a mobile phone, a tablet computer, a smart Bluetooth device, a laptop computer, a personal computer and other devices; the server can be a single server or a server cluster composed of multiple servers.

[0161] For example, in this embodiment, the method of the embodiment of the present application is described in detail by taking the object analysis device specifically integrated in the terminal as an example. This embodiment provides an object analysis device, such as Figure 4 As shown, the object analysis device may include:

[0162] An acquisition module 401 is used to acquire multiple target object groups to be analyzed, and to acquire a distribution indication parameter of the target object groups on at least one portrait tag, wherein the distribution indication parameter is used to indicate the distribution quantity of the target object groups on at least one tag value to which the portrait tag belongs;

[0163] A distribution determination module 402, configured to determine the distribution degree of the target object group on the portrait tag based on the distribution indication parameter of the target object group on the portrait tag;

[0164] A vector determination module 403 is used to determine the distribution vectors corresponding to the at least two target object groups respectively based on the distribution indication parameters of the at least two target object groups on the target label when there are at least two target object groups distributed on the same target label;

[0165] The situation determination module 404 is used to obtain distribution difference information of the at least two target object groups on the target tags based on the distribution vectors respectively corresponding to the at least two target object groups.

[0166] In some embodiments, the acquisition module 401 is specifically used for:

[0167] Obtaining basic information and behavior information of multiple users through a preset portrait system, and performing information processing based on the basic information and behavior information of the multiple users, and assigning at least one portrait tag and a tag value corresponding to the user on the portrait tag to each of the multiple users;

[0168] Get multiple object circle selection tags and the circle selection tag values ​​corresponding to the above object circle selection tags;

[0169] Based on the plurality of object circle selection tags, the circle selection tag values ​​corresponding to the object circle selection tags, and the portrait tags corresponding to the plurality of users, the plurality of users are divided into groups to obtain a plurality of target object groups to be analyzed.

[0170] In some embodiments, the distribution determination module 402 is specifically used to:

[0171] Determine the probability corresponding to the target object group on at least one label value based on the distribution quantity of the target object group on at least one label value to which the portrait label belongs;

[0172] Based on the probability of the target object group corresponding to at least one label value, the distribution degree of the target object group on the portrait label is determined.

[0173] In some embodiments, the determination of the distribution degree of the target object group on the portrait label based on the probability corresponding to the target object group on at least one label value is specifically used for:

[0174] Based on the probability corresponding to the target object group on at least one label value, the information entropy of the target object group on the portrait label is calculated, and the information entropy is determined as the distribution degree of the target object group on the portrait label.

[0175] In some embodiments, the object analysis device further includes a group division module, and the group division module is specifically used to:

[0176] If the information entropy is greater than or equal to the first preset information threshold, the target object group is divided into groups based on the portrait label to which the information entropy belongs.

[0177] In some embodiments, the at least two target object groups include a first object group and a second object group, and the vector determination module 403 is specifically used to:

[0178] Constructing a coordinate system based on at least one target tag value to which the target tag belongs, wherein each target tag value serves as a coordinate axis in the coordinate system;

[0179] Determine a first vector corresponding to the first object group in the coordinate system based on the distribution quantity corresponding to the first object group on at least one of the target label values;

[0180] Based on the distribution quantity corresponding to the second object group on at least one of the target label values, a second vector corresponding to the second object group in the coordinate system is determined.

[0181] In some embodiments, the situation determination module 404 is specifically used to:

[0182] Angle information between the first vector and the second vector is calculated, and distribution difference information between the first object group and the second object group on the target label is determined based on the angle information.

[0183] From the above, it can be seen that by obtaining multiple target object groups to be analyzed, and obtaining the distribution indication parameters of the above target object groups on at least one portrait label, the above distribution indication parameters are used to indicate the distribution quantity of the target object group on at least one label value belonging to the above portrait label; based on the distribution indication parameters of the above target object group on the above portrait label, the distribution degree of the above target object group on the above portrait label is determined; when there are at least two target object groups distributed on the same target label, based on the distribution indication parameters of the at least two target object groups on the above target label, the distribution vectors corresponding to the at least two target object groups are determined; based on the distribution vectors corresponding to the at least two target object groups, the distribution difference information of the at least two target object groups on the target label is obtained, thereby improving the operation effect through the distribution degree of a single target object group on the portrait label, and the distribution difference information between multiple target object groups on the target label.

[0184] Accordingly, the embodiment of the present application also provides an electronic device, which may be a terminal, and the terminal may be a smart phone, a tablet computer, a laptop computer, a touch screen, a game console, a personal computer (PC, Personal Computer), a personal digital assistant (Personal Digital Assistant, PDA) and other terminal devices. Figure 5 As shown, Figure 5A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device 500 includes a processor 501 having one or more processing cores, a memory 502 having one or more computer-readable storage media, and a computer program stored in the memory 502 and executable on the processor. The processor 501 is electrically connected to the memory 502. Those skilled in the art will appreciate that the electronic device structure shown in the figure does not constitute a limitation on the electronic device, and may include more or fewer components than shown, or combine certain components, or arrange components differently.

[0185] The processor 501 is the control center of the electronic device 500. It uses various interfaces and lines to connect various parts of the entire electronic device 500, executes various functions of the electronic device 500 and processes data by running or loading software programs and / or modules stored in the memory 502, and calling data stored in the memory 502, thereby monitoring the electronic device 500 as a whole.

[0186] In the embodiment of the present application, the processor 501 in the electronic device 500 will load instructions corresponding to the processes of one or more application programs into the memory 502 according to the following steps, and the processor 501 will run the application programs stored in the memory 502 to implement various functions:

[0187] Acquire multiple target object groups to be analyzed, and acquire distribution indication parameters of the target object groups on at least one portrait tag, wherein the distribution indication parameters are used to indicate the distribution quantity of the target object groups on at least one tag value to which the portrait tag belongs;

[0188] Determining the distribution degree of the target object group on the portrait label based on the distribution indication parameter of the target object group on the portrait label;

[0189] When there are at least two target object groups distributed on the same target label, determining distribution vectors corresponding to the at least two target object groups respectively based on distribution indication parameters of the at least two target object groups on the target label;

[0190] Based on the distribution vectors respectively corresponding to the at least two target object groups, distribution difference information of the at least two target object groups on the target labels is obtained.

[0191] In some embodiments, the step of obtaining a plurality of target object groups to be analyzed includes:

[0192] Obtaining basic information and behavior information of multiple users through a preset portrait system, and performing information processing based on the basic information and behavior information of the multiple users, and assigning at least one portrait tag and a tag value corresponding to the user on the portrait tag to each of the multiple users;

[0193] Get multiple object circle selection tags and the circle selection tag values ​​corresponding to the above object circle selection tags;

[0194] Based on the plurality of object circle selection tags, the circle selection tag values ​​corresponding to the object circle selection tags, and the portrait tags corresponding to the plurality of users, the plurality of users are divided into groups to obtain a plurality of target object groups to be analyzed.

[0195] In some embodiments, the determining the distribution degree of the target object group on the portrait tag based on the distribution indication parameter of the target object group on the portrait tag includes:

[0196] Determine the probability corresponding to the target object group on at least one label value based on the distribution quantity of the target object group on at least one label value to which the portrait label belongs;

[0197] Based on the probability of the target object group corresponding to at least one label value, the distribution degree of the target object group on the portrait label is determined.

[0198] In some embodiments, the determining of the distribution degree of the target object group on the portrait label based on the probability corresponding to the target object group on at least one label value includes:

[0199] Based on the probability corresponding to the target object group on at least one label value, the information entropy of the target object group on the portrait label is calculated, and the information entropy is determined as the distribution degree of the target object group on the portrait label.

[0200] In some embodiments, it also includes:

[0201] If the information entropy is greater than or equal to the first preset information threshold, the target object group is divided into groups based on the portrait label to which the information entropy belongs.

[0202] In some embodiments, the at least two target object groups include a first object group and a second object group, and the determining of the distribution vectors corresponding to the at least two target object groups respectively based on the distribution indication parameters of the at least two target object groups on the target tags includes:

[0203] Constructing a coordinate system based on at least one target tag value to which the target tag belongs, wherein each target tag value serves as a coordinate axis in the coordinate system;

[0204] Determine a first vector corresponding to the first object group in the coordinate system based on the distribution quantity corresponding to the first object group on at least one of the target label values;

[0205] Based on the distribution quantity corresponding to the second object group on at least one of the target label values, a second vector corresponding to the second object group in the coordinate system is determined.

[0206] In some embodiments, obtaining the distribution difference information of the at least two target object groups on the target tags based on the distribution vectors respectively corresponding to the at least two target object groups includes:

[0207] Angle information between the first vector and the second vector is calculated, and distribution difference information between the first object group and the second object group on the target label is determined based on the angle information.

[0208] Therefore, the electronic device 500 provided in this embodiment can bring the following technical effects: improving operational effects.

[0209] The specific implementation of the above operations can be found in the previous embodiments, which will not be described in detail here.

[0210] Optional, such as Figure 5 As shown, the electronic device 500 further includes: a touch screen 503, a radio frequency circuit 504, an audio circuit 505, an input unit 506, and a power supply 507. The processor 501 is electrically connected to the touch screen 503, the radio frequency circuit 504, the audio circuit 505, the input unit 506, and the power supply 507, respectively. Those skilled in the art can understand that Figure 5 The electronic device structure shown in the figure does not constitute a limitation of the electronic device, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.

[0211] The touch display screen 503 can be used to display a graphical user interface and receive operation instructions generated by the user acting on the graphical user interface. The touch display screen 503 may include a display panel and a touch panel. Among them, the display panel may be used to display information input by the user or information provided to the user and various graphical user interfaces of the electronic device, and these graphical user interfaces may be composed of graphics, text, icons, videos and any combination thereof. Optionally, the display panel may be configured in the form of a liquid crystal display (LCD, Liquid Crystal Display), an organic light-emitting diode (OLED, Organic Light-Emitting Diode) and the like. The touch panel may be used to collect the user's touch operation on or near it (such as the user using any suitable object or attachment such as a finger, a stylus, etc. on the touch panel or near the touch panel), and generate corresponding operation instructions, and the operation instructions execute corresponding programs. Optionally, the touch panel may include two parts, a touch detection device and a touch controller. Among them, the touch detection device detects the user's touch orientation, detects the signal brought by the touch operation, and transmits the signal to the touch controller; the touch controller receives the touch information from the touch detection device, converts it into the touch point coordinates, and then sends it to the processor 501, and can receive the command sent by the processor 501 and execute it. The touch panel can cover the display panel. When the touch panel detects a touch operation on or near it, it is transmitted to the processor 501 to determine the type of touch event, and then the processor 501 provides a corresponding visual output on the display panel according to the type of touch event. In an embodiment of the present application, the touch panel and the display panel can be integrated into the touch display screen 503 to realize the input and output functions. However, in some embodiments, the touch panel and the touch panel can be used as two independent components to realize the input and output functions. That is, the touch display screen 503 can also be used as a part of the input unit 506 to realize the input function.

[0212] The radio frequency circuit 504 may be used to send and receive radio frequency signals, so as to establish wireless communication with a network device or other electronic devices through wireless communication, and to send and receive signals between the network device or other electronic devices.

[0213] The audio circuit 505 can be used to provide an audio interface between the user and the electronic device through a speaker and a microphone. The audio circuit 505 can transmit the electrical signal converted from the received audio data to the speaker, which is converted into a sound signal for output; on the other hand, the microphone converts the collected sound signal into an electrical signal, which is received by the audio circuit 505 and converted into audio data, and then the audio data is output to the processor 501 for processing, and then sent to another electronic device through the radio frequency circuit 504, or the audio data is output to the memory 502 for further processing. The audio circuit 505 may also include an earplug jack to provide communication between an external headset and an electronic device.

[0214] The input unit 506 may be used to receive input numbers, character information or user feature information (such as fingerprint, iris, facial information, etc.), and generate keyboard, mouse, joystick, optical or trackball signal input related to user settings and function control.

[0215] The power supply 507 is used to supply power to various components of the electronic device 500. Optionally, the power supply 507 can be logically connected to the processor 501 through a power management system, so that the power management system can manage charging, discharging, and power consumption. The power supply 507 can also include one or more DC or AC power supplies, recharging systems, power failure detection circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0216] although Figure 5 Not shown, the electronic device 500 may also include a camera, a sensor, a wireless fidelity module, a Bluetooth module, etc., which will not be described in detail here.

[0217] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0218] A person of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by instructions, or by controlling related hardware through instructions. The instructions may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0219] To this end, an embodiment of the present application provides a computer-readable storage medium, in which multiple computer programs are stored, and the computer program can be loaded by a processor to execute the steps in any object analysis method provided in the embodiment of the present application. For example, the computer program can execute the following steps:

[0220] Acquire multiple target object groups to be analyzed, and acquire distribution indication parameters of the target object groups on at least one portrait tag, wherein the distribution indication parameters are used to indicate the distribution quantity of the target object groups on at least one tag value to which the portrait tag belongs;

[0221] Determining the distribution degree of the target object group on the portrait label based on the distribution indication parameter of the target object group on the portrait label;

[0222] When there are at least two target object groups distributed on the same target label, determining distribution vectors corresponding to the at least two target object groups respectively based on distribution indication parameters of the at least two target object groups on the target label;

[0223] Based on the distribution vectors respectively corresponding to the at least two target object groups, distribution difference information of the at least two target object groups on the target labels is obtained.

[0224] In some embodiments, the step of obtaining a plurality of target object groups to be analyzed includes:

[0225] Obtaining basic information and behavior information of multiple users through a preset portrait system, and performing information processing based on the basic information and behavior information of the multiple users, and assigning at least one portrait tag and a tag value corresponding to the user on the portrait tag to each of the multiple users;

[0226] Get multiple object circle selection tags and the circle selection tag values ​​corresponding to the above object circle selection tags;

[0227] Based on the plurality of object circle selection tags, the circle selection tag values ​​corresponding to the object circle selection tags, and the portrait tags corresponding to the plurality of users, the plurality of users are divided into groups to obtain a plurality of target object groups to be analyzed.

[0228] In some embodiments, the determining the distribution degree of the target object group on the portrait tag based on the distribution indication parameter of the target object group on the portrait tag includes:

[0229] Determine the probability corresponding to the target object group on at least one label value based on the distribution quantity of the target object group on at least one label value to which the portrait label belongs;

[0230] Based on the probability of the target object group corresponding to at least one label value, the distribution degree of the target object group on the portrait label is determined.

[0231] In some embodiments, the determining of the distribution degree of the target object group on the portrait label based on the probability corresponding to the target object group on at least one label value includes:

[0232] Based on the probability corresponding to the target object group on at least one label value, the information entropy of the target object group on the portrait label is calculated, and the information entropy is determined as the distribution degree of the target object group on the portrait label.

[0233] In some embodiments, it also includes:

[0234] If the information entropy is greater than or equal to the first preset information threshold, the target object group is divided into groups based on the portrait label to which the information entropy belongs.

[0235] In some embodiments, the at least two target object groups include a first object group and a second object group, and the determining of the distribution vectors corresponding to the at least two target object groups respectively based on the distribution indication parameters of the at least two target object groups on the target tags includes:

[0236] Constructing a coordinate system based on at least one target tag value to which the target tag belongs, wherein each target tag value serves as a coordinate axis in the coordinate system;

[0237] Determine a first vector corresponding to the first object group in the coordinate system based on the distribution quantity corresponding to the first object group on at least one of the target label values;

[0238] Based on the distribution quantity corresponding to the second object group on at least one of the target label values, a second vector corresponding to the second object group in the coordinate system is determined.

[0239] In some embodiments, obtaining the distribution difference information of the at least two target object groups on the target tags based on the distribution vectors respectively corresponding to the at least two target object groups includes:

[0240] Angle information between the first vector and the second vector is calculated, and distribution difference information between the first object group and the second object group on the target label is determined based on the angle information.

[0241] It can be seen that the computer program can be loaded by the processor to execute the steps in any object analysis method provided in the embodiments of the present application, thereby bringing the following technical effects: improving operational results.

[0242] The specific implementation of the above operations can be found in the previous embodiments, which will not be described in detail here.

[0243] The computer-readable storage medium may include: a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0244] Since the computer program stored in the computer-readable storage medium can execute the steps in any object analysis method provided in the embodiments of the present application, the beneficial effects that can be achieved by any object analysis method provided in the embodiments of the present application can be achieved. Please refer to the previous embodiments for details and will not be repeated here.

[0245] The above is a detailed introduction to an object analysis method, device, electronic device and computer-readable storage medium provided in the embodiments of the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for technical personnel in this field, according to the ideas of the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. An object analysis method, characterized in that: The method comprises: Acquire multiple target object groups to be analyzed that participate in the same operation activity, and acquire a distribution indication parameter of the target object groups on at least one portrait tag, wherein the distribution indication parameter is used to indicate the distribution quantity of the target object group on at least one tag value to which the portrait tag belongs, and the portrait tags of at least two of the target object groups are different; Determining the distribution degree of the target object group participating in the operation activity on the portrait tag based on the distribution indication parameter of the target object group on the portrait tag; When there are at least two target object groups distributed on the same target label, a coordinate system is constructed based on at least one target label value to which the target label belongs, wherein each target label value serves as a coordinate axis in the coordinate system, and the target label is the same portrait label associated with the at least two target object groups, and the at least two target object groups include a first object group and a second object group; Determine, based on the distribution quantity corresponding to the first object group on at least one of the target label values, a coordinate value of the first object group on a coordinate axis corresponding to the target label value, and determine, based on the coordinate value, a first vector corresponding to the first object group in the coordinate system; Determine, based on the distribution quantity corresponding to the second object group on at least one of the target label values, a coordinate value of the second object group on a coordinate axis corresponding to the target label value, and determine, based on the coordinate value, a second vector corresponding to the second object group in the coordinate system; Based on the first vector and the second vector, distribution difference information of the at least two target object groups participating in the operation activity on the target label is obtained.

2. The object analysis method according to claim 1, characterized in that: The obtaining of multiple target object groups to be analyzed that participate in the same operation activity includes: Obtaining basic information and behavior information of multiple users through a preset portrait system, and performing information processing based on the basic information and behavior information of the multiple users, and assigning at least one portrait tag and a tag value corresponding to the user on the portrait tag to each of the multiple users; Obtain multiple object circle selection tags and circle selection tag values ​​corresponding to the object circle selection tags; Based on the plurality of object circle selection tags, the circle selection tag values ​​corresponding to the object circle selection tags, and the portrait tags corresponding to the plurality of users, the plurality of users are divided into groups to obtain a plurality of target object groups to be analyzed.

3. The object analysis method according to claim 1, characterized in that: The determining, based on the distribution indication parameter of the target object group on the portrait tag, the distribution degree of the target object group participating in the operation activity on the portrait tag comprises: Determining the probability corresponding to the target object group on at least one label value based on the distribution quantity of the target object group on at least one label value to which the portrait label belongs; Based on the probability of the target object group corresponding to at least one label value, the distribution degree of the target object group on the portrait label is determined.

4. The object analysis method according to claim 3, characterized in that: The determining, based on the probability corresponding to the target object group on at least one label value, the distribution degree of the target object group on the portrait label comprises: Based on the probability corresponding to the target object group on at least one label value, the information entropy of the target object group on the portrait label is calculated, and the information entropy is determined as the distribution degree of the target object group on the portrait label.

5. The object analysis method according to claim 4, characterized in that: Also includes: If the information entropy is greater than or equal to a first preset information threshold, the target object group is divided into groups based on the portrait label to which the information entropy belongs.

6. The object analysis method according to claim 1, characterized in that: The obtaining, based on the first vector and the second vector, distribution difference information of the at least two target object groups participating in the operation activity on the target label includes: Angle information between the first vector and the second vector is calculated, and distribution difference information between the first object group and the second object group on the target tag is determined based on the angle information.

7. An object analysis device, characterized in that: The device comprises: An acquisition module, used to acquire multiple target object groups to be analyzed that participate in the same operation activity, and to acquire a distribution indication parameter of the target object groups on at least one portrait tag, wherein the distribution indication parameter is used to indicate the distribution quantity of the target object group on at least one tag value to which the portrait tag belongs, and the portrait tags of at least two of the target object groups are different; A distribution determination module, configured to determine the distribution degree of the target object group participating in the operation activity on the portrait tag based on the distribution indication parameter of the target object group on the portrait tag; A vector determination module, for constructing a coordinate system based on at least one target label value to which the target label belongs when there are at least two target object groups distributed on the same target label, wherein each of the target label values ​​serves as a coordinate axis in the coordinate system, and the target label is the same portrait label associated with the at least two target object groups, and the at least two target object groups include a first object group and a second object group; based on the distribution quantity corresponding to the first object group on at least one of the target label values, determining the coordinate value of the first object group on the coordinate axis corresponding to the target label value, and determining the first vector corresponding to the first object group in the coordinate system based on the coordinate value; based on the distribution quantity corresponding to the second object group on at least one of the target label values, determining the coordinate value of the second object group on the coordinate axis corresponding to the target label value, and determining the second vector corresponding to the second object group in the coordinate system based on the coordinate value; A situation determination module is used to obtain distribution difference information of the at least two target object groups participating in the operation activity on the target label based on the first vector and the second vector.

8. An electronic device, characterized in that: It comprises a processor and a memory, wherein the memory stores a plurality of instructions; the processor loads instructions from the memory to execute the steps in the object analysis method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor to execute the steps in the object analysis method according to any one of claims 1 to 6.

Images