A method and device for selecting a target object

Abstract

The present invention discloses a method and device for selecting a target object, and relates to the field of computer technology. A specific implementation of the method includes: obtaining a sample user set, dividing the sample users therein into two categories, one category constituting a positive sample user set, and the other category constituting a negative sample user set; determining the number of preset behaviors of each of the positive sample users or the negative sample users for each object; determining the weight of each of the objects according to the number; and selecting a target object from the objects according to the weight. This implementation achieves the technical effect that the selected target object is more in line with expectations and actual conditions.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method, device, electronic device and computer-readable medium for selecting a target object. Background Art

[0002] In the big data environment, data volume accumulates rapidly. In order to analyze the value contained in the massive data, it is very important to filter out the valuable data. Data screening plays a vital role in the entire data processing process. The purpose of data screening is to improve the availability of previously collected and stored related data, which is more conducive to later data analysis.

[0003] In the process of implementing the present invention, the inventors found that there are at least the following problems in the prior art:

[0004] The increase in data volume and the complexity of the structure make it inevitable that data screening for big data will consume more resources. Therefore, it is necessary to select the appropriate algorithm in order to quickly and accurately screen out valuable data. Each algorithm has its own environment for use. The increase in data complexity in the big data environment increases the difficulty of selecting a suitable algorithm. Secondly, the huge amount of big data also makes it increasingly difficult to analyze valuable data through a single algorithm, and the accuracy of selecting data that meets the expected target needs to be improved. Summary of the invention

[0005] In view of this, an embodiment of the present invention provides a method and device for selecting a target object, which can determine the weight of the target object according to the classification results of the sample user. The target object selected according to the weight of the object is more accurate; the classification model constructed using the target object as a training sample is more in line with expectations and actual conditions.

[0006] To achieve the above-mentioned purpose, according to one aspect of an embodiment of the present invention, a method for selecting a target object is provided, comprising: obtaining a sample user set, dividing the sample users therein into two categories, one category constituting a positive sample user set, and the other category constituting a negative sample user set; determining the number of preset behaviors of each of the positive sample users or the negative sample users for each object; determining the weight of each of the objects according to the number; and selecting a target object from the objects according to the weight.

[0007] Optionally, after selecting the target object from the objects according to the weight, the method further includes: determining the number of preset behaviors of each of the sample users for each of the target objects; constructing a feature vector according to the number of preset behaviors of each of the sample users for each of the target objects; and constructing a classification model according to the feature vector, wherein the classification model is used to classify users.

[0008] Optionally, each of the objects C is determined according to the quantity i The weight V(C i ) include:

[0009]

[0010]

[0011]

[0012] Among them, the positive sample user or the negative sample user X n For each of the objects C i The number of preset behaviors is M(X n |C i ) times, each user Y q For each of the objects C i The number of preset behaviors is M(Y q |C i ) times; n=1,2,3,…,N1, N1 is the number of the positive sample users or the negative sample users X; i=1,2,3,…,N2, N2 is the number of the objects C; q=1,2,3,…,N3, N3 is the number of all users Y.

[0013] Optionally, the method of obtaining a sample user set and dividing the sample users therein into two categories, one category constituting a positive sample user set and the other category constituting a negative sample user set includes: obtaining a sample user set and dividing the sample users therein into two categories on average, one category constituting a positive sample user set and the other category constituting a negative sample user set.

[0014] Optionally, the method of constructing a classification model based on the feature vector includes: constructing a first classification model M1 based on the feature vector using a random forest algorithm; constructing a second classification model M2 based on the feature vector using a gradient boosting tree algorithm; then the classification model M is: M=λM1+(1-λ)M2, where λ is a preset coefficient and λ is not greater than 1.

[0015] To achieve the above-mentioned purpose, according to another aspect of an embodiment of the present invention, there is provided a device for selecting a target object, comprising: a classification module, used to obtain a sample user set, and divide the sample users therein into two categories, one category constituting a positive sample user set, and the other category constituting a negative sample user set; a calculation module, used to determine the number of preset behaviors of each of the positive sample users or the negative sample users for each object; an analysis module, used to determine the weight of each of the objects according to the number; and a selection module, used to select a target object from the objects according to the weight.

[0016] Optionally, the calculation module is also used to determine the number of preset behaviors of each sample user for each target object; the device also includes: a feature module, used to construct a feature vector according to the number of preset behaviors of each sample user for each target object; and a model module, used to construct a classification model according to the feature vector, wherein the classification model is used to classify users.

[0017] Optionally, each of the objects C is determined according to the quantity i The weight V(C i ) include:

[0018]

[0019]

[0020]

[0021] The number of preset behaviors of the positive sample user or the negative sample user X for each object C is M(X n |C i ) times, each user Y q For each of the objects C i The number of preset behaviors is M(Y q |C i ) times; n=1,2,3,…,N1, N1 is the number of the positive sample users or the negative sample users X; i=1,2,3,…,N2, N2 is the number of the objects C; q=1,2,3,…,N3, N3 is the number of all users Y.

[0022] Optionally, the classification module is further used to obtain a sample user set, and divide the sample users therein into two categories on average, one category constituting a positive sample user set, and the other category constituting a negative sample user set.

[0023] Optionally, the model module is also used to construct a first classification model M1 based on the feature vector using a random forest algorithm; and to construct a second classification model M2 based on the feature vector using a gradient boosting tree algorithm; then the classification model M is: M=λM1+(1-λ)M2, where λ is a preset coefficient and λ is not greater than 1.

[0024] To achieve the above-mentioned purpose, according to another aspect of an embodiment of the present invention, there is provided an electronic device, comprising: one or more processors; a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement any of the methods described in a method for selecting a target object.

[0025] To achieve the above objective, according to another aspect of an embodiment of the present invention, a computer-readable medium is provided, on which a computer program is stored, and when the program is executed by one or more processors, any method described in a method for selecting a target object is implemented.

[0026] An embodiment of the above invention has the following advantages or beneficial effects: because a technical means is adopted to calculate the weight of an object according to the classification of sample users, the technical problems of excessive data volume and low accuracy when selecting a target object are overcome, thereby achieving the technical effect that the selected target object is more in line with expectations and actual conditions.

[0027] The further effects of the above-mentioned non-conventional optional manner will be described below in conjunction with the specific implementation manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The accompanying drawings are used to better understand the present invention and do not constitute an improper limitation of the present invention.

[0029] Figure 1 is a schematic diagram of main steps of a method for selecting a target object according to an embodiment of the present invention;

[0030] Figure 2 is a schematic diagram of the main parts of a device for selecting a target object according to an embodiment of the present invention;

[0031] Figure 3 is an exemplary system architecture diagram to which embodiments of the present invention may be applied;

[0032] Figure 4 It is a schematic diagram of the structure of a computer system of a terminal device or a server suitable for implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0033] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0034] Figure 1 is a schematic diagram of the main steps of a method for selecting a target object according to an embodiment of the present invention. Figure 1 As shown:

[0035] Step S101 indicates obtaining a sample user set, and dividing the sample users therein into two categories, one of which constitutes a positive sample user set and the other constitutes a negative sample user set. The purpose of this step is to narrow the scope of data analysis and improve the accuracy of selecting the target object. Further, the sample users in the sample user set are evenly divided into two categories, one of which constitutes a positive sample user set and the other constitutes a negative sample user set. The classification standard can be determined based on the association between the sample users and the target object. The purpose is to ensure the number of sample users, form a comparison mechanism, and improve the accuracy of target object selection.

[0036] For example, a sample user set is obtained. When the target object is associated with the gender of the sample user, all sample users are divided into two categories according to gender, one category constitutes a positive sample (male sample) user set, and the other category constitutes a negative sample (female sample) user set, and the target object is selected according to the user's gender characteristics.

[0037] Step S102 indicates determining the number of preset behaviors of each of the positive sample users or the negative sample users for each object. The purpose of this step is to further narrow the analysis scope of the data and improve the accuracy of selecting the target object.

[0038] For example, in the field of e-commerce, when the object is a certain category of goods, buyers are selected as the sample user set, and the preset behaviors are browsing, following, adding to shopping cart and placing orders. The number of views, following, adding to shopping cart and ordering for each category of goods by each positive sample user or negative sample user is determined.

[0039] Step S103 indicates determining the weight of each object according to the quantity. The purpose of this step is to analyze and quantify the quantity of the preset behaviors to calculate the weight.

[0040] Determine each of the objects C according to the quantity i The weight V(C i ) include:

[0041]

[0042]

[0043]

[0044] Among them, the positive sample user or the negative sample user X n For each of the objects C i The number of preset behaviors is M(X n |C i ) times, each user Y q For each of the objects C i The number of preset behaviors is M(Yq |C i )Second-rate;

[0045] n=1,2,3,…,N1, where N1 is the number of the positive sample users or the negative sample users X;

[0046] i=1, 2, 3, ..., N2, N2 is the number of objects C;

[0047] q=1,2,3,…,N3, where N3 is the number of all users Y.

[0048] Represents all positive sample users or negative sample users for object C i The sum of the number of preset behaviors; Represents the sample user X n The sum of the number of preset behaviors for all objects; Represents all users (including all sample users) for object C i The sum of the number of preset behaviors; Indicates user Y q The sum of the number of preset behaviors for all objects.

[0049] When there is more than one preset behavior, for example, the number of the first preset behavior is M1, the number of the second preset behavior is M2, and the number of the sth preset behavior is M s , s=1,2,3,…N4, N4 is the number of preset behaviors M, then:

[0050]

[0051] Step S104 represents selecting a target object from the objects according to the weight. For example, a weight threshold can be preset, and objects with weight values ​​greater than or less than the threshold are selected as target objects; or a preset number can be set, and the objects are sorted in descending order according to the weight value, and the first preset number of objects are selected as target objects.

[0052] After selecting the target object from the objects according to the weight, the method may further include: determining the number of preset behaviors of each sample user for each target object; constructing a feature vector according to the number of preset behaviors of each sample user for each target object; constructing a classification model according to the feature vector, wherein the classification model is used to classify users. The purpose of this step is to construct a classification model using the target object as a training sample.

[0053] For example, in the field of e-commerce, select N categories of goods as target objects, determine the number of orders (i.e., preset behaviors) placed by the i-th positive sample user and the i-th negative sample user (i.e., buyer) for each target object, and denote them as F 1i , F 2i , F 3i , …, F Ni , construct the feature vector T i :

[0054] T i =(F 1i ,F 2i ,F 3i ,…,F Ni )

[0055] Use the relevant algorithm to solve the binary classification problem, with the feature vector T i As the input value, a decision tree model can be constructed by machine learning with feature vectors as training samples.

[0056] Furthermore, the method for constructing a classification model based on the feature vector includes: based on the feature vector, using a random forest algorithm (a classification algorithm including multiple decision trees) to construct a first classification model M1; based on the feature vector, using a gradient boosting tree algorithm (also called a multiple additive regression tree, an iterative decision tree algorithm, which consists of multiple decision trees, and the conclusions of all trees are accumulated to make the final decision result) to construct a second classification model M2; then the classification model M is: M=λM1+(1-λ)M2, where λ is a preset coefficient, and λ is not greater than 1.

[0057] The purpose of this step is to integrate the models of random forest and gradient boosting tree algorithms to further improve the classification accuracy and generalization ability of the model.

[0058] Figure 2 is a schematic diagram of the main parts of a device 200 for selecting a target object according to an embodiment of the present invention. Figure 2 As shown:

[0059] The classification module 201 is used to obtain a sample user set and classify the sample users therein into two categories, one of which constitutes a positive sample user set and the other constitutes a negative sample user set; the purpose is to narrow the analysis scope of the data and improve the accuracy of selecting the target object.

[0060] The classification module 201 can also be used to obtain a sample user set and divide the sample users therein into two categories, one of which constitutes a positive sample user set and the other constitutes a negative sample user set. The classification standard can be determined according to the association relationship between the sample users and the target object. The purpose is to ensure the number of sample users, form a comparison mechanism, and improve the accuracy of target object selection.

[0061] For example, a sample user set is obtained. When the target object is associated with the gender of the sample user, all sample users are divided into two categories according to gender, one category constitutes a positive sample (male sample) user set, and the other category constitutes a negative sample (female sample) user set, and the target object is selected according to the user's gender characteristics.

[0062] The calculation module 202 is used to determine the number of preset behaviors of each of the positive sample users or the negative sample users for each object; the purpose is to further narrow the analysis scope of the data and improve the accuracy of selecting the target object.

[0063] For example, in the field of e-commerce, when the object is a certain category of goods, buyers are selected as the sample user set, and the preset behaviors are browsing, following, adding to shopping cart and placing orders. The number of views, following, adding to shopping cart and ordering for each category of goods by each positive sample user or negative sample user is determined.

[0064] The analysis module 203 is used to determine the weight of each of the objects according to the quantity; the purpose is to analyze and quantify the quantity of the preset behaviors, so as to calculate the weight.

[0065] Determine each of the objects C according to the quantity i The weight V(C i ) include:

[0066]

[0067]

[0068]

[0069] Among them, the positive sample user or the negative sample user X n For each of the objects C i The number of preset behaviors is M(X n |C i ) times, each user Y q For each of the objects C i The number of preset behaviors is M(Y q |C i) times; n=1,2,3,…,N1, N1 is the number of the positive sample users or the negative sample users X; i=1,2,3,…,N2, N2 is the number of the objects C; q=1,2,3,…,N3, N3 is the number of all users Y.

[0070] Represents all positive sample users or negative sample users for object C i The sum of the number of preset behaviors; Represents sample user X n The sum of the number of preset behaviors for all objects; Represents all users (including all sample users) for object C i The sum of the number of preset behaviors; Indicates user Y q The sum of the number of preset behaviors for all objects.

[0071] When there is more than one preset behavior, for example, the number of the first preset behavior is M1, the number of the second preset behavior is M2, and the number of the sth preset behavior is M s , s=1,2,3,…N4, N4 is the number of preset behaviors M, then:

[0072]

[0073] The selection module 204 is used to select the target object from the objects according to the weight. For example, a weight threshold can be preset, and objects with weight values ​​greater than or less than the threshold can be selected as target objects; or a preset number can be set, and the objects are sorted in descending order according to the weight value, and the first preset number of objects are selected as target objects.

[0074] The calculation module 202 is also used to determine the number of preset behaviors of each sample user for each target object; the device 200 may also include: a feature module, which is used to construct a feature vector according to the number of preset behaviors of each sample user for each target object; a model module, which is used to construct a classification model according to the feature vector, wherein the classification model is used to classify users. The purpose is to use the target object as a training sample to construct a classification model.

[0075] For example, in the field of e-commerce, select N categories of goods as target objects, determine the number of orders (i.e., preset behaviors) placed by the i-th positive sample user and the i-th negative sample user (i.e., buyer) for each target object, and denote them as F 1i , F 2i , F 3i , …, F Ni , construct the feature vector T i :

[0076] T i =(F 1i ,F 2i ,F 3i ,…,F Ni )

[0077] Use the relevant algorithm to solve the binary classification problem, with the feature vector T i As the input value, a decision tree model can be constructed by machine learning with feature vectors as training samples.

[0078] The model module is also used to construct a first classification model M1 according to the feature vector using a random forest algorithm (a classification algorithm including multiple decision trees); and to construct a second classification model M2 according to the feature vector using a gradient boosting tree algorithm (also called multiple additive regression tree, an iterative decision tree algorithm, which consists of multiple decision trees, and the conclusions of all trees are accumulated to make a final decision result);

[0079] Then the classification model M is: M=λM1+(1-λ)M2, where λ is a preset coefficient and λ is not greater than 1.

[0080] The purpose is to integrate the models of random forest and gradient boosting tree algorithms to further improve the classification accuracy and generalization ability of the model.

[0081] Figure 3 An exemplary system architecture 300 is shown to which a method for selecting a target object or an apparatus for selecting a target object according to an embodiment of the present invention can be applied.

[0082] like Figure 3 As shown, the system architecture 300 may include terminal devices 301, 302, 303, a network 304 and a server 305. The network 304 is used to provide a medium for communication links between the terminal devices 301, 302, 303 and the server 305. The network 304 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0083] The user can use the terminal devices 301, 302, 304 to interact with the server 305 through the network 304 to receive or send messages, etc. Various communication client applications can be installed on the terminal devices 301, 302, 303, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0084] The terminal devices 301 , 302 , and 303 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, and desktop computers, etc.

[0085] The server 305 may be a server that provides various services, such as a backend management server that provides support for shopping websites browsed by users using the terminal devices 301, 302, and 303. The backend management server may analyze and process the received data such as product information query requests, and feed back the processing results (such as target push information and product information) to the terminal device.

[0086] It should be noted that the method for selecting a target object provided in the embodiment of the present invention is generally executed by the server 305 , and accordingly, a device for selecting a target object is generally disposed in the server 305 .

[0087] It should be understood that Figure 3 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0088] Figure 4 Shown is a schematic diagram of the structure of a computer system 400 of a terminal device suitable for implementing an embodiment of the present invention. Figure 4 The terminal device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0089] like Figure 4 As shown, the computer system 400 includes a central processing unit (CPU) 401, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 402 or a program loaded from a storage part 408 into a random access memory (RAM) 403. In the RAM 403, various programs and data required for the operation of the system 400 are also stored. The CPU 401, the ROM 402, and the RAM 403 are connected to each other via a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0090] The following components are connected to the I / O interface 405: an input section 406 including a keyboard, a mouse, etc.; an output section 407 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 408 including a hard disk, etc.; and a communication section 409 including a network interface card such as a LAN card, a modem, etc. The communication section 409 performs communication processing via a network such as the Internet. A drive 410 is also connected to the I / O interface 405 as needed. A removable medium 411, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 410 as needed, so that a computer program read therefrom is installed into the storage section 408 as needed.

[0091] In particular, according to the embodiments disclosed in the present invention, the process described in the above step diagram can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the step diagram. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 409, and / or installed from the removable medium 411. When the computer program is executed by the central processing unit (CPU) 401, the above-mentioned functions defined in the system of the present invention are executed.

[0092] It should be noted that the computer-readable medium shown in the present invention includes a computer-readable signal medium or a computer-readable storage medium, or any combination of the above two. Computer-readable storage media include, but are not limited to, electrical, magnetic, optical, electromagnetic, infrared, semiconductor systems, devices or devices, or any combination of the above. Computer-readable storage media specifically include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any combination of the above. In the present invention, a computer-readable storage medium includes any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device or device; a computer-readable signal medium includes a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code, and such a propagated data signal can take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any combination of the above signals. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF (radio frequency), etc., or any combination of the above media.

[0093] The step diagram or block diagram in the accompanying drawings illustrates the possible implementation architecture, functions and operations of the system, method and computer program product according to various embodiments of the present invention. Each box in the step diagram or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of the code contains one or more executable instructions for implementing the specified logical function. It should be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed in parallel, and sometimes they can be executed in the opposite order, and the execution order depends on the functions involved. It should also be noted that each box in the block diagram or step diagram and its combination can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0094] The modules or units involved in the embodiments of the present invention may be implemented in software or hardware. The modules or units described may also be provided in a processor, for example, may be described as: a processor including a classification module, a calculation module, an analysis module and a selection module. The names of these modules or units do not, in certain cases, constitute limitations on the modules or units themselves, for example, the analysis module may also be described as "a module for determining the weight of each of the objects according to the quantity".

[0095] On the other hand, an embodiment of the present invention further provides a computer-readable medium, which may be included in the device described in the above embodiment; or may exist independently without being assembled into the device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by a device, the device includes: obtaining a sample user set, dividing the sample users therein into two categories, one category constituting a positive sample user set, and the other category constituting a negative sample user set; determining the number of preset behaviors of each of the positive sample users or the negative sample users for each object; determining the weight of each of the objects according to the number; and selecting a target object from the objects according to the weight.

[0096] According to the technical solution of the embodiment of the present invention, a technical means is adopted to calculate the weight of the object according to the classification of sample users, thereby overcoming the technical problems of excessive data volume and low accuracy when selecting target objects, thereby achieving the technical effect that the selected target objects are more in line with expectations and actual conditions.

[0097] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for selecting a target object, characterized in that: include: Obtain a sample user set, and divide the sample users therein into two categories, one of which constitutes a positive sample user set, and the other constitutes a negative sample user set; Determine the number of preset behaviors of each positive sample user in the positive sample user set or each negative sample user in the negative sample user set for each object; Determining the weight of each of the objects according to the number includes: calculating the weight of each of the objects using the number of preset behaviors of all sample users in the sample user set for each object and the number of preset behaviors of each positive sample user for each object; or calculating the weight of each of the objects using the number of preset behaviors of all sample users in the sample user set for each object and the number of preset behaviors of each negative sample user for each object; selecting a target object from the objects according to the weight; Wherein, each object C is determined according to the quantity i The weight V(C i ) include: Among them, the positive sample user or the negative sample user X n For each of the objects C i The number of preset behaviors is M(X n |C i ) times, each user Y q For each of the objects C i The number of preset behaviors is M(Y q |C i )Second-rate; n=1,2,3,…,N1, where N1 is the number of the positive sample users or the negative sample users X; i=1, 2, 3, ..., N2, N2 is the number of objects C; q=1,2,3,…,N3, where N3 is the number of all users Y.

2. The method according to claim 1, characterized in that: After selecting the target object from the objects according to the weight, the method further includes: Determining the number of preset behaviors of each of the sample users for each of the target objects; Constructing a feature vector according to the number of preset behaviors of each sample user for each target object; A classification model is constructed according to the feature vector, wherein the classification model is used to classify users.

3. The method according to claim 1, characterized in that A method for obtaining a sample user set and dividing the sample users therein into two categories, one of which constitutes a positive sample user set and the other constitutes a negative sample user set, includes: A sample user set is obtained, and the sample users are evenly divided into two categories, one of which constitutes a positive sample user set and the other constitutes a negative sample user set.

4. The method according to claim 2, characterized in that: The method for constructing a classification model according to the feature vector includes: According to the feature vector, a first classification model M1 is constructed using a random forest algorithm; According to the feature vector, a second classification model M2 is constructed using a gradient boosting tree algorithm; Then the classification model M is: M=λM1+(1-λ)M2, where λ is a preset coefficient and λ is not greater than 1.

5. A device for selecting a target object, characterized in that: include: The classification module is used to obtain a sample user set and divide the sample users into two categories, one of which constitutes a positive sample user set and the other constitutes a negative sample user set; A calculation module, used to determine the number of preset behaviors of each positive sample user in the positive sample user set or each negative sample user in the negative sample user set for each object; The analysis module is used to determine the weight of each of the objects according to the number, including: calculating the weight of each of the objects using the number of preset behaviors of all sample users in the sample user set for each object and the number of preset behaviors of each positive sample user for each object; or calculating the weight of each of the objects using the number of preset behaviors of all sample users in the sample user set for each object and the number of preset behaviors of each negative sample user for each object; A selection module, used for selecting a target object from the objects according to the weight; Wherein, each object C is determined according to the quantity i The weight V(C i ) include: Among them, the positive sample user or the negative sample user X n For each of the objects C i The number of preset behaviors is M(X n |C i ) times, each user Y q For each of the objects C i The number of preset behaviors is M(Y q |C i )Second-rate; n=1,2,3,…,N1, where N1 is the number of the positive sample users or the negative sample users X; i=1, 2, 3, ..., N2, N2 is the number of objects C; q=1,2,3,…,N3, where N3 is the number of all users Y.

6. The device according to claim 5, characterized in that The calculation module is also used to determine the number of preset behaviors of each of the sample users for each of the target objects; The device also includes: A feature module, used for constructing a feature vector according to the number of preset behaviors of each sample user for each target object; The model module is used to construct a classification model according to the feature vector, wherein the classification model is used to classify users.

7. The device according to claim 5, characterized in that The classification module is further used to obtain a sample user set and divide the sample users therein into two categories on average, one category constituting a positive sample user set and the other constituting a negative sample user set.

8. The device according to claim 6, characterized in that The model module is also used to construct a first classification model M1 using a random forest algorithm according to the feature vector; According to the feature vector, a second classification model M2 is constructed using a gradient boosting tree algorithm; Then the classification model M is: M=λM1+(1-λ)M2, where λ is a preset coefficient and λ is not greater than 1.

9. An electronic device, characterized in that: include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 4.

10. A computer readable medium having a computer program stored thereon, characterized in that: When the program is executed by one or more processors, the method according to any one of claims 1 to 4 is implemented.

Images