KR20180116953A - Supporting apparatus for advertisement and control method thereof, and providing apparatus for advertisement - Google Patents

Abstract

The present invention proposes an advertisement supporting apparatus, an operating method thereof, and an advertisement providing apparatus, capable of expecting optimum advertisement efficiency by selectively applying matching information (feature) for matching between an advertisement and a user in a user behavior log collected in association with an advertisement selection, in association with a machine learning-based matching algorithm, and further improving return on investment (ROI) of an advertiser.

Description

TECHNICAL FIELD [0001] The present invention relates to an advertisement supporting apparatus, an operation method thereof, and an advertisement providing apparatus,

The present invention relates to a method for selectively applying matching information for matching between an advertisement and a user in a user behavior log collected in association with an advertisement selection in connection with a machine learning based matching algorithm for matching between an advertisement and a user .

As the market for applications running on mobile devices such as smart phones and tablet pads is rapidly expanding, the related advertising market is rapidly expanding.

In the mobile device (hereinafter referred to as a 'terminal'), when an application corresponding to an advertisement medium is executed or a service page (for example, a web page) is accessed, an advertisement provided from the advertisement providing device is received, And can access the advertiser page linked to the advertisement when the displayed advertisement is clicked.

On the other hand, in the advertisement providing device, it is general to adopt an advertisement targeting technique to selectively provide the advertisement suitable for the user in order to enhance the advertisement effect.

As an example of the above advertisement targeting technique, there is a method of matching an advertisement with a user using a learning algorithm based on a machine learning. This has the advantage that the efficiency of the advertisement can be optimized in a limited resource.

Here, the matching algorithm based on machine learning refers to a method of analyzing and modeling supervised learning, that is, piled training data (user behavior log). Usually, the modeling is performed through various methods such as linear regression and logistic regression And the matching rate between the advertisement and the user can be obtained as a prediction value coming out through modeling.

Thus, in the matching algorithm based on the machine learning, the prediction value corresponding to the matching rate between the advertisement and the user is obtained, and the advertisement is selected based on the result value having the highest prediction value, You can do it.

On the other hand, in the above-mentioned machine learning based matching algorithm, an error between the prediction value and the actual value decreases as the number of matching information applied for matching between the advertisement and the user increases.

However, in the matching algorithm based on the machine learning, modeling is performed in order to minimize the error based only on the given training data (user behavior log), resulting in a new problem of over fitting.

As a result, in order to optimize the efficiency of the advertisement through the existing machine learning based matching algorithm, it is important to select and apply the matching information applied for matching between the advertisement and the user.

SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and it is an object of the present invention to provide a method and apparatus for matching a user- And a method for selectively applying matching information for matching between an advertisement and a user.

According to another aspect of the present invention, there is provided an advertisement support apparatus comprising: a processor for performing an operation related to matching an advertisement and a user; And a memory in which at least one instruction executed by the processor is stored, the at least one instruction being for each of a plurality of user behavior logs collected in connection with the advertisement selection, after each user behavior log is identified A calculating step of calculating a degree of correlation with the advertisement selection based on the number of times the advertisement is selected as the number of times the advertisement is selected; And selecting at least a part of the user behavior log of the plurality of user behavior logs according to a priority having a high degree of correlation with the advertisement selection and determining the selected user behavior log as matching information for matching the advertisement and the user And a decision command.

More specifically, the degree of correlation with the advertisement selection is a value obtained by normalizing the advertisement selection count of each of the plurality of user behavior logs using an average value and a variance value of the number of advertisement selection of each of the plurality of user behavior logs And the user action log having the number of advertisement selection times larger than the average value is calculated as a higher value than the remaining user action log.

More specifically, the plurality of user behavior logs may include at least one of a first user behavior log collected during a first period and a second user behavior log collected during a second period neighboring the first period Wherein the decision command is to select at least a part of the user behavior logs having a high degree of correlation between the first user behavior log and the second user behavior log as the matching information, .

More specifically, the decision command determines the optimal number of matching information through the learned matching algorithm using the training data including the user behavior log, and determines the user behavior log of the optimum number of matching information as the matching information .

More specifically, when the test data is applied to the matching algorithm in place of the training data, the optimum number of matching information is a number of users in the test data corresponding to a point at which the matching rate between the advertisement and the user is derived from the matching algorithm, And the number of action logs.

According to an aspect of the present invention, there is provided an advertisement providing apparatus including: a processor for performing an operation related to providing an advertisement; And a memory in which at least one instruction executed via the processor is stored, wherein the at least one instruction is adapted to cause at least a portion of a plurality of user activity logs collected in connection with a particular user & As matching information for matching the acquired information; And a selection command for selecting an advertisement matched with the specific user based on the matching information and providing the selected advertisement to the specific user.

More specifically, the matching information is at least a part of the user behavior log selected from among the plurality of user behavior logs according to a priority having a high degree of correlation with the advertisement selection, And the number of times the advertisement is selected after the behavior log is verified.

According to another aspect of the present invention, there is provided a method of operating an advertisement receiver management apparatus, the method comprising: after each user action log is verified for each of a plurality of user action logs collected in association with an advertisement selection, A calculation step of calculating a degree of correlation with the advertisement selection based on a number of times of selection of an advertisement which is a selected number of times; And a step of selecting at least a part of the user behavior logs of the plurality of user behavior logs according to a priority having a high degree of correlation with the advertisement selection and determining the selected user behavior log as matching information for matching an advertisement and a user The method comprising the steps of:

More specifically, the degree of correlation with the advertisement selection is a value obtained by normalizing the advertisement selection count of each of the plurality of user behavior logs using an average value and a variance value of the number of advertisement selection of each of the plurality of user behavior logs And the user action log having the number of advertisement selection times larger than the average value is calculated as a higher value than the remaining user action log.

More specifically, the plurality of user behavior logs may include at least one of a first user behavior log collected during a first period and a second user behavior log collected during a second period neighboring the first period Wherein the determining step comprises the step of selecting at least a part of the user behavior logs having a high degree of correlation between the first user behavior log and the second user behavior log as the matching information, do.

More specifically, the determining step determines the optimal number of matching information through the learned matching algorithm using the training data including the user behavior log, determines the user behavior log for the optimum number of matching information as the matching information .

More specifically, when the test data is applied to the matching algorithm in place of the training data, the optimum number of matching information is a number of users in the test data corresponding to a point at which the matching rate between the advertisement and the user is derived from the matching algorithm, And the number of action logs.

Accordingly, with respect to the advertisement supporting apparatus, the operation method thereof, and the advertisement providing apparatus according to the embodiment of the present invention, in relation to the machine learning based matching algorithm, among the user behavior logs collected in association with the advertisement selection, By applying the matching information for matching, it is possible to expect an optimal advertisement efficiency and further improve the ROI (Return On Investment) of the advertiser.

1 is a schematic configuration diagram of an advertisement providing system according to an embodiment of the present invention;
2 is a schematic configuration diagram of a terminal according to an embodiment of the present invention;
3 is an exemplary diagram of a hardware system for implementing a terminal according to an embodiment of the present invention.
4 is a schematic configuration diagram of an advertisement supporting apparatus according to an embodiment of the present invention;
5 is an exemplary diagram illustrating a matching rate in a machine learning based matching algorithm according to an embodiment of the present invention;
FIG. 6 illustrates an example of a hardware system for implementing an advertisement supporting apparatus according to an embodiment of the present invention; FIG.
7 is a schematic configuration diagram of an advertisement providing apparatus according to an embodiment of the present invention;
8 is an exemplary diagram of a hardware system for implementing an advertisement providing apparatus according to an embodiment of the present invention;
FIG. 9 is a schematic flowchart for explaining an operation flow in an advertisement providing system according to an embodiment of the present invention; FIG.
FIG. 10 is a schematic flowchart for explaining an operation flow in an advertisement supporting apparatus according to an embodiment of the present invention; FIG.
11 is a schematic flowchart for explaining an operational flow in an advertisement providing apparatus according to an embodiment of the present invention;

It is noted that the technical terms used herein are used only to describe specific embodiments and are not intended to limit the invention. It is also to be understood that the technical terms used herein are to be interpreted in a sense generally understood by a person skilled in the art to which the present invention belongs, Should not be construed to mean, or be interpreted in an excessively reduced sense. Further, when a technical term used herein is an erroneous technical term that does not accurately express the spirit of the present invention, it should be understood that technical terms that can be understood by a person skilled in the art are replaced. In addition, the general terms used in the present invention should be interpreted according to a predefined or prior context, and should not be construed as being excessively reduced.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings, wherein like reference numerals refer to like or similar elements throughout the several views, and redundant description thereof will be omitted. In the following description, well-known functions or constructions are not described in detail since they would obscure the invention in unnecessary detail. It is to be noted that the accompanying drawings are only for the purpose of facilitating understanding of the present invention, and should not be construed as limiting the scope of the present invention with reference to the accompanying drawings. The spirit of the present invention should be construed as extending to all modifications, equivalents, and alternatives in addition to the appended drawings.

Hereinafter, an embodiment of the present invention will be described with reference to the accompanying drawings.

1 is a diagram illustrating an advertisement providing system according to an embodiment of the present invention.

1, an advertisement providing system according to an embodiment of the present invention, a terminal 10 for receiving an advertisement, an advertisement supporting apparatus 20 for determining matching information for matching between an advertisement and a user, And an advertisement providing device 30 for providing an advertisement to the mobile terminal 10.

In the case of the advertisement providing system according to the embodiment of the present invention, in addition to the above-described configuration, an advertiser apparatus (not shown) for providing an advertiser page (web page) to which the terminal 10, As shown in FIG.

The terminal 10 refers to a user's mobile device capable of performing a series of operations such as receiving an advertisement through an advertisement medium and accessing a service page through selection of an advertisement.

For example, the terminal 10 may be a smart phone, a portable terminal, a mobile terminal, a personal digital assistant (PDA), a portable multimedia player (PMP) A telematics terminal, a navigation terminal, a personal computer, a notebook computer, a slate PC, a tablet PC, an ultrabook, a wearable device, (Including a Smartwatch, a Smart Glass, and a HMD (Head Mounted Display)), a Wibro terminal, an IPTV (Internet Protocol Television) terminal, a smart TV, a digital broadcasting terminal, a television A 3D television, a home theater system, an AVN (Audio Video Navigation) terminal, an A / V (Audio / Video) system, a flexible terminal, and the like. Any device capable of receiving advertisements via any advertising medium may be included.

Here, as an advertisement medium enabling the reception of advertisements in the terminal 10, for example, an application executed in the terminal 10 itself or a service page (for example, a web page) accessible by calling the browser from the terminal 10 .

Such an advertisement medium includes an SDK (Software Development Kit) related to the advertisement reception in the course of its development. As a result, the terminal 10 can receive a series of advertisements through the SDK included in the advertisement medium, Operation can be performed.

For reference, the advertising medium may support the collection of user behavior logs performed in the ad-supported device 20, as well as the above-described series of functions leading to ad request and reception and display.

Here, the user behavior log includes, for example, a CPS (Campaign Sequence), an ADT (AD Type), a BLT (Bill Type), a TGS (Target Sequence), an MDS (Media Sequence), a CCB (Eg, heathy, music, sport), user id, conversion rate, ad request and reception, and display and selection, service page (eg, web page, advertiser page) And all information related to the user's behavior through the terminal 10 such as user behavior information (e.g., purchase, browse, inquiry, search).

Accordingly, in the terminal 10, for example, advertisement reception and display and advertisement selection are performed in the application or service page. When the advertiser page is accessed when the advertisement is selected or the user accesses the other service page, To the advertisement support apparatus 20 as an action log.

In addition, the terminal 10 according to the embodiment of the present invention can receive an advertisement through a pull method of requesting and receiving an advertisement to the advertisement providing device 30 using its device identification information, It is needless to say that the advertisement can be received through a push method in which an advertisement provided by the advertisement providing apparatus 30 is received without a separate advertisement request for the advertisement providing apparatus 30. [

The advertisement supporting device 20 refers to a data management platform for determining matching information for matching between an advertisement and a user and is a data management platform in which a user behavior log collected (received) Log analysis and analysis to classify ad recipient groups.

The advertisement providing apparatus 30 refers to an advertisement platform for providing advertisements to the terminal 10 and provides advertisements by referring to the matching information determined by the advertisement supporting apparatus 20. [

As described above, each of the advertisement supporting apparatus 20 and the advertisement providing apparatus 30 adopted as a system configuration in the embodiment of the present invention can be implemented in the form of a web server, a database server, a proxy server, Load balancing mechanism, and various software that allows the service device to operate on the Internet or other network, and may be implemented as a computerized system. The network may also be an http network and may be a private line, an intranet, or any other network, and the connection between each of the configurations in the ad serving system according to an embodiment of the present invention may be any data And may be connected to a secure network so as not to be attacked by a hacker or other third party.

Further, each of the ad-supported device 20 and the advertisement providing device 30 may include a plurality of database servers, and the database server may be connected to the advertisement supporting device (not shown) via any type of network connection including the distributed database server architecture 20 and the advertisement providing apparatus 30, respectively.

Meanwhile, the advertisement supporting apparatus 20 according to an embodiment of the present invention determines matching information applied for matching between advertisements and users in a matching algorithm based on machine learning.

In general, machine learning based matching algorithm learns Supervised learning, that is, a large number of accumulated user activity logs as training data, and applies some of the user behavior logs as matching information, Can be derived as a prediction result.

In this case, the error between the predicted value and the actual value decreases as the number of matching information for matching between the advertisement and the user increases. On the other hand, in the matching algorithm based on the existing machine learning, in order to minimize the error based on the given training data As the modeling proceeds, a new problem of overfitting occurs.

Here, overfitting refers to the state of the model, which can not distinguish between noise and signal, and uses excessive noise for learning, thereby reducing prediction performance of future data.

As a conventional method for overfitting, there is a method of deriving a prediction result that is more suitable for a real situation by reducing the degree of matching information through regularization. In order to formulate such a prediction, There is a required limit.

Accordingly, in an embodiment of the present invention, a new method for selectively applying matching information for matching between an advertisement and a user in a user behavior log collected in association with an advertisement selection in association with a matching algorithm based on a machine learning Hereinafter, each configuration in the advertisement providing system for realizing this will be described in more detail.

FIG. 2 shows a configuration of a terminal 10 according to an embodiment of the present invention.

2, the terminal 10 according to an embodiment of the present invention may have a configuration including an app advertiser 11 and a web advertiser 12.

The application advertisement unit 11 refers to a configuration for performing a series of operations related to an advertisement when an application is executed in the terminal 10, which can be understood as an SDK (Software Development Kit) embedded in an application.

The web advertisement unit 12 is another advertisement medium existing in the terminal 10 and is configured to perform a series of operations related to an advertisement when a service page (for example, a web page) is accessed. Browser. ≪ / RTI >

Here, when accessing a service page (web page), the browser can perform a series of operations related to the advertisement through a method of reading the source code of the SDK inserted in the service page.

As described above, in the case of a browser serving as the web advertisement unit 12, an OS-based browser provided by an operating system (OS) of the terminal 10, or an application itself provided by an application executed in the terminal 10 Based browser.

Here, an application capable of providing a browser by itself can be generally called a hybrid application (e.g., a mobile exclusive shopping mall application).

All or at least a part of the configuration of the terminal 10 including the above-mentioned application advertisement unit 11 and the web advertisement unit 12 may be implemented in the form of a software module or a hardware module executed by the processor, A combination of a module and a hardware module may be implemented.

The terminal 10 according to an embodiment of the present invention includes a display unit 13 for displaying advertisements, an input / output unit 14 for supporting input / output functions such as display and selection of advertisements, And a communication unit 15 for supporting communication with the advertisement supporting apparatus 20 and the advertisement providing apparatus 30, respectively.

The configurations of the display unit 13, the input / output unit 14 and the communication unit 15 are the same as those of the display unit 1321, the input / output unit 1320, and the communication unit 1310 And the detailed description thereof will be described below.

The terminal 10 according to the embodiment of the present invention receives and displays the advertisement from the advertisement providing device 30 through the core configuration including the application advertisement part 11 and the web advertisement part 12, It is possible to access a separate advertiser page provided by the advertiser device (not shown), and the user behavior log according to the series of processes can be transmitted to the advertisement support device 20. Hereinafter, Each configuration in the terminal 10 will be described in detail.

The application advertisement unit 11 performs a function of receiving an advertisement.

More specifically, the application advertisement unit 11 requests an advertisement to the advertisement providing apparatus 30 according to the execution of the application. When the advertisement is received from the advertisement providing apparatus 30 according to the advertisement request, The received advertisement is displayed in the designated advertisement display area.

At this time, the application advertisement unit 11 can request the advertisement providing apparatus 30 to advertise using the device identification information (e.g., MDN, IMEI) of the terminal 10 according to the execution of the application.

In addition, the app advertising unit 11 performs a function of requesting a connection to an advertiser device (not shown).

More specifically, after the advertisement received from the advertisement providing device 30 is displayed in the designated advertisement display area within the application execution screen, if the click of the advertisement is confirmed, A connection to the advertiser apparatus (not shown) is requested to the web advertisement unit 12 corresponding to the browser in the terminal 10 according to an embodiment.

In this case, the selected advertisement may include address information of an advertiser device (not shown). If the click of the advertisement is confirmed, the app advertiser 11 transmits the address information included in the advertisement to the web advertisement unit 12 to request access to an advertiser device (not shown).

In addition, after the advertisement advertisement unit 11 displays the advertisement received from the advertisement providing device 30 in the advertisement display area in the application execution screen, the connection to the advertiser device (not shown) It is possible to transmit the user behavior log to the advertisement support apparatus 20.

Here, the method of delivering the user behavior log may be based on, for example, a real-time delivery method, or may be based on a batch transmission method of collectively transmitting accumulated user action logs for a predetermined time.

The web advertisement unit 12 performs a function of accessing a service page.

More specifically, when a click on the in-application advertisement is confirmed through the application advertisement unit 11, the web advertisement unit 12 transmits the advertisement information to the advertiser device (Not shown), thereby making it possible to use various services (for example, purchase of goods) provided by the advertiser apparatus (not shown).

On the other hand, the web advertisement unit 12 is an advertisement medium different from the application advertisement unit 11, and is a configuration capable of performing a series of operations related to an advertisement according to a service page (for example, a web page) connection.

The web advertisement unit 12 may request an advertisement to the advertisement providing apparatus 30 according to a connection to a service page (for example, a web page) separately from the application advertisement unit 11, The advertisement can be displayed in the advertisement display area on the service page.

The web advertisement unit 12 displays the advertisement received from the advertisement providing apparatus 30 in the advertisement display area of the service page as in the case of the application advertisement unit 11 and then the click of the advertisement is confirmed , Various services (e.g., merchandise sales) provided by the advertiser device (not shown) can also be used by accessing the advertiser device (not shown) using the address information embedded in the advertisement.

The web advertisement unit 12 displays the advertisement received from the advertisement providing apparatus 300 in the advertisement display area in the service page as in the case of the application advertisement unit 11, (Not shown) is connected, the user behavior log may be transmitted to the advertisement support apparatus 20.

In each of the configurations of the terminal 10 described above, it is noted that the present invention can be implemented in the form of a software module or a hardware module executed by a processor, or a combination of a software module and a hardware module.

As such, a form of software module, hardware module, or combination of software module and hardware module executed by a processor may be implemented in a real hardware system (e.g., a computer system).

Therefore, a hardware system 1000 for implementing the terminal 10 according to an embodiment of the present invention will be described with reference to FIG.

For reference, it should be noted that the following description is an example for implementing each configuration in the terminal 10 described above in the hardware system 1000, and that each configuration and operation thereof may be different from an actual system something to do.

3, a hardware system 1000 according to an embodiment of the present invention includes a processor unit 1100, a memory interface unit 1200, and a peripheral device interface unit 1300 .

Each such configuration in the hardware system 1000 may be a separate component or integrated into one or more integrated circuits, and each of these components may be coupled by a bus system (not shown).

Here, in the case of a bus system, any one or more individual physical buses, communication lines / interfaces, and / or multi-drop or point-to-point links connected by appropriate bridges, adapters, and / And abstraction that represents point-to-point connections.

The processor unit 1100 executes various software modules stored in the memory unit 1210 by communicating with the memory unit 1210 through the memory interface unit 1200 to perform various functions in the hardware system .

2 may be stored in the memory unit 1210 in the form of a software module. In addition, the memory 1210 may store the application advertisement unit 11 and the web advertisement unit 12, which are components of the terminal 10, (OS) may be additionally stored.

In the case of an operating system (for example, an embedded operating system such as I-OS, Android, Darwin, RTXC, LINUX, UNIX, OS X, WINDOWS, or VxWorks), typical system tasks (such as memory management, , Power management, and the like), a set of instructions, a software component, and / or a driver, and facilitates communication between various hardware modules and software modules.

For reference, the memory unit 1210 may include a memory hierarchy including, but not limited to, a cache, a main memory, and a secondary memory, such as RAM (e.g., SRAM, (E.g., DRAM, DDRAM), ROM, FLASH, magnetic and / or optical storage devices such as disk drives, magnetic tape, compact discs, and digital video discs.

The peripheral device interface unit 1300 serves to enable communication between the processor unit 1100 and a peripheral device.

Here, in the case of a peripheral device, the hardware system 1000 is provided for providing different functions. In an embodiment of the present invention, for example, a communication unit 1310 and an input / output unit 1320 may be included.

Here, the communication unit 1310 performs a function of providing a communication function with another device. For example, an antenna system, an RF transceiver, one or more amplifiers, a tuner, one or more oscillators, a digital signal processor, CODEC) chipset, and memory, and the like, and may include known circuits that perform this function.

Examples of the communication protocol supported by the communication unit 1310 include a wireless LAN (WLAN), a DLNA (Digital Living Network Alliance), a Wireless Broadband (Wibro), a World Interoperability for Microwave Access ), Code Division Multiple Access (CDMA), Code Division Multi Access (CDMA2000), Enhanced Voice-Data Optimized or Enhanced Voice-Only (EV-DO), Wideband CDMA (WCDMA) , HSDPA (High Speed Downlink Packet Access), HSUPA (High Speed Uplink Packet Access), IEEE 802.16, Long Term Evolution (LTE), Long Term Evolution-Advanced (LTE-A) Wireless Mobile Broadband Service (WMBS), Bluetooth, Radio Frequency Identification (RFID), Infrared Data Association (IrDA), Ultra-Wideband (UWB), ZigBee, Near Field Communication NFC, Ultra Sound Communication (USC), visible light Shin: could include (Visible Light Communication VLC), Wi Fi (Wi-Fi), Wi Fi Direct (Wi-Fi Direct). The wired communication network may be a wired LAN, a wired WAN, a power line communication (PLC), a USB communication, an Ethernet, a serial communication, Cables, and the like, and may include any protocol that can provide a communication environment with other devices, which is not limited at present.

In addition, the input / output unit 1320 serves as a controller for controlling an I / O device interlocked with other hardware systems. In an embodiment of the present invention, an application execution screen display, an advertisement selection (click) And can control the display unit 1321 that supports display of the page connection screen and the like.

Here, it goes without saying that the display unit 1321 supports a touch interface for confirming an advertisement selection (click).

The display unit 1321 may be a liquid crystal display (LCD), a thin film transistor LCD (TFT-LCD), a light emitting diode (LED), an organic light emitting diode OLED, Organic LED, AMOLED, Active Matrix OLED, Retina Display, Flexible Display, and 3 Dimension Display.

As a result, each configuration in the terminal 10, which is stored in the memory unit 1210 in the form of a software module according to the above configuration, is stored in the memory interface unit 1200 and the peripheral device 1010 in the form of a command executed by the processor unit 1100. [ The communication unit 1310 and the input / output unit 1320 communicate with each other via the interface unit 1300 to display advertisements received from the advertisement providing apparatus 30 and to access advertiser pages. And transmits the user action log to the advertisement support apparatus 20.

The description of the terminal 10 according to the embodiment of the present invention will be described below.

It is assumed that the user activity log related to the advertisement selection (click) is collected in the advertisement support device 20 prior to the specific description of the advertisement support device 20.

Here, a plurality of user behavior logs collected according to an embodiment of the present invention may include a first user behavior log collected during a first period (e.g., three months), a second user behavior log collected during a second period adjacent to the first period And a second user activity log collected over a period of time (e.g., 3 months).

FIG. 4 shows a schematic configuration of an advertisement supporting apparatus 20 according to an embodiment of the present invention.

4, the advertisement supporting apparatus 20 according to an embodiment of the present invention includes a calculating unit 21 for calculating a degree of correlation with an advertisement selection (click) of a user behavior log, A determination unit 22 for selectively determining the information as matching information, and a providing unit 23 for providing matching information.

All or at least a part of the configuration of the advertisement supporting apparatus 20 including the calculating unit 21, the determining unit 22 and the providing unit 23 may be implemented in the form of a software module executed by a processor or a hardware module Or software modules and hardware modules may be implemented in a combined form.

Here, the software module can be understood as, for example, a command executed by a processor that controls an operation in the advertisement supporting apparatus 20, and this command can be a form embedded in a memory in the advertisement supporting apparatus 20 There will be.

The advertisement supporting apparatus 20 according to an embodiment of the present invention may further include a communication unit 24 for supporting communication between the terminal 10 and the advertisement providing apparatus 30 in addition to the above- Lt; / RTI >

For reference, the configuration of the communication unit 24 corresponds to the configuration of the communication unit 2310 to be described below with reference to FIG. 6, and a detailed description thereof will be given below.

The advertisement supporting apparatus 20 according to an embodiment of the present invention may be configured to perform matching for matching between advertisements and users through a core configuration including a calculating unit 21, a determining unit 22, In the following, each configuration in the advertisement support device 20 for realizing this will be described in detail.

The calculating unit 21 calculates the degree of correlation with the advertisement selection (Click).

More specifically, the calculating unit 21 calculates the degree of correlation with the advertisement selection for each of the plurality of user action logs collected in association with the advertisement selection.

Here, the relation with the advertisement selection means that the selection of the advertisement has been substantially performed after each user action log has been confirmed.

At this time, the calculating unit 21 calculates the correlation with the advertisement selection for each of the plurality of user behavior logs collected in association with the advertisement selection using the number of the advertisement selection, which is the number of times the advertisement is selected after each user action log is confirmed Can be calculated.

That is, the calculation unit 21 calculates an average value and a variance value for the number of times of the advertisement selection of each of a plurality of user behavior logs, calculates a mean value and a variance value using a mean value and a variance value, The degree of correlation is calculated through the value (Z) obtained by normalizing the number of times of selection of each advertisement of the action log.

[Equation 1]

Z = (X -?) /?

Here, 'X' represents the number of advertisement selection in each user behavior log, 'μ' represents an average value of the number of advertisement selection, and 'σ' represents a dispersion value of the number of advertisement selection.

In this case, the large average value of the number of advertisement selection means that the values are gathered at the center, and it can be understood that the error occurrence probability with respect to the predicted value is low at the time of modeling through the matching algorithm, It can be understood that the error occurrence probability with respect to the predicted value is high when modeling through the matching algorithm.

As a result, the degree of correlation with the advertisement selection can be calculated through the value (Z) obtained by normalizing the number of times of selection of the advertisement according to the above [Equation 1], and the greater the value Z normalized to the number of selection of the advertisement, It can be interpreted that the degree is high.

On the other hand, as mentioned above, the variance value for the number of advertisement selection is directly related to the error occurrence probability for the prediction value at the time of modeling through the matching algorithm.

In order to reflect the stability, the calculation unit 21 calculates a variance value (Z) for the value (Z) obtained by normalizing each advertisement selection number before the user behavior log is determined as the matching information for reflecting in the matching algorithm The following equation [2] is reflected.

[Equation 2]

Correlation (indicator) = value (Z) normalized number of advertisement selection / variance value

As a result, the result of adding the variance value to the value (Z) obtained by normalizing the number of advertisement selection times through the above equation (2) can be finally calculated as a correlation for each user action log.

The determination unit 22 performs a function of determining matching information.

More specifically, when the calculation of the degree of correlation with the advertisement selection for each of the plurality of collected user log information is completed, the determination unit 22 determines that at least some of the user behavior logs have a high priority with respect to the advertisement selection And the selected user action log is determined as matching information for matching the advertisement and the user.

In this regard, a plurality of pieces of user log information collected in accordance with an embodiment of the present invention includes a first user behavior log collected in the first period and a second user behavior log collected in the second period adjacent to the first period As well.

Accordingly, the determination unit 22 integrates the first user behavior log and the second user behavior log, and selects at least a part of the integrated user behavior logs according to the priority with high correlation with the advertisement selection, As matching information for matching.

The determination of the matching information by integrating the first user behavior log and the second user behavior log is performed in order to consider that the variance value in the first period and the second period may vary.

That is, the user behavior log collected in the first period and the user behavior logs collected in the second period have a difference not only in the average value but also in the variance value.

Therefore, if the variance value in the first period is large even if the user selects the number of times of advertisement selection in the first period, the degree of correlation with the advertisement selection will be calculated to be low. If the number of advertisement selection is large in the second period It is possible to calculate the degree of correlation with the advertisement selection higher than the user behavior log with the number of advertisement selection times in the first period if the variance value in the second period is small.

As a result, the decision unit 22 applies a method of determining matching information by combining two or more periods (first period and second period) in which variance values may be different from each other, thereby determining a user action The log can be selected as the matching information.

Before determining the matching information, the determining unit 22 confirms the matching information optimum number through the learned matching algorithm based on the training data corresponding to the user behavior log, The user behavior logs corresponding to the optimum number of pieces of matching information identified in the log are selected according to the correlation degree priorities and are determined as matching information.

Here, when the test data is applied to the matching algorithm instead of the training data, the optimal number of matching information is the number of user behavior logs in the test data corresponding to the point at which the matching rate between the advertisement and the user is derived from the matching algorithm .

In this regard, FIG. 5 shows a change in the matching rate between the user and the advertisement according to the number of user behavior logs applied in the machine learning based matching algorithm according to an embodiment of the present invention.

Here, (A) represents a change in the matching rate when the number of user behavior logs corresponding to the training data increases as matching information, (B) shows that the tester data is applied to the matching algorithm instead of the training data, And a change in the matching rate when the number of user action logs in the data increases as matching information.

In this case, in (B), which indicates the change in the matching rate when new tester data is applied to the matching algorithm instead of the training data, it can be seen that the matching rate is lowered at a specific time point when the number of user action logs in the tester data is increased. The phenomenon that the matching rate decreases at a specific point in time when the number of user action logs is increased can be understood as overfitting.

This overfitting can be overcome by reducing the number of user behavior logs applied as matching information to the matching algorithm. Thus, in the present invention, the matching rate is highest from the matching algorithm to the user. In other words, The number of user behavior logs in the test data corresponding to the time point before the decrease in the number of user behavior logs can be confirmed as the optimum number of matching information.

The providing unit 23 performs a function of providing matching information.

More specifically, when the user behavior logs corresponding to the optimum number of matching information among a plurality of collected user behavior logs are determined as matching information according to the correlation degree priority, the providing unit 23 transmits the determined matching information to the advertisement providing apparatus 30), the advertisement providing apparatus 30 reflects the received matching information to the matching algorithm to process the matching between the advertisement and the terminal 10 user, and as a result of the processing, To the terminal (10).

Meanwhile, in each of the configurations of the advertisement supporting apparatus 20 described above, it is noted that the present invention can be implemented in the form of a software module or a hardware module executed by a processor, or a combination of a software module and a hardware module have.

As such, a form of software module, hardware module, or combination of software module and hardware module executed by a processor may be implemented in a real hardware system (e.g., a computer system).

6, a description will now be made of a hardware system 2000 implementing each configuration in the advertisement supporting apparatus 20 according to an embodiment of the present invention.

On the other hand, among the configurations of the hardware system 2000 shown in Fig. 6, configurations in which the names coincide with those in the hardware system 1000 described with reference to Fig. 3 can be regarded as the same configuration, A specific example or description of the same configuration will be omitted.

6, a hardware system 2000 for implementing an advertisement supporting apparatus 20 according to an embodiment of the present invention includes a processor unit 2100, a memory interface unit 2200, (2300). ≪ / RTI >

The processor unit 2100 communicates with the memory unit 2210 through the memory interface unit 2200 to execute various software modules stored in the memory unit 2210 to perform various functions in the hardware system .

In the memory unit 2210, the calculating unit 21, the determining unit 22, and the providing unit 23, which are components of the advertisement supporting apparatus 20 described with reference to FIG. 4, And other operating systems can be additionally stored.

The peripheral device interface unit 2300 serves to enable communication between the processor unit 2100 and the peripheral device.

Here, in the case of a peripheral device, the hardware system 2000 is provided for providing different functions. In an embodiment of the present invention, for example, a communication unit 2310 may be included.

Here, the communication unit 2310 performs a function of providing a communication function with other devices. For example, an antenna system, an RF transceiver, one or more amplifiers, a tuner, one or more oscillators, a digital signal processor, a codec CODEC) chipset, and memory, and the like, and may include known circuits that perform this function.

As a result, each configuration in the advertisement supporting apparatus 20 stored in the memory unit 2210 in the form of a software module is stored in the memory interface unit 2100 and the peripheral device interface unit 2100 in the form of a command executed by the processor unit 2100. [ The matching information for matching between the advertisement and the user can be determined by processing the interface with the communication unit 2310 via the communication unit 2300.

Hereinafter, with reference to FIG. 6, each configuration of the advertisement supporting apparatus 20 will be described in connection with the hardware system 2000. FIG.

The calculating unit 21 calculates the degree of correlation with the advertisement selection (Click).

More specifically, the calculating unit 21 calculates a degree of correlation with the advertisement selection for each of the plurality of user behavior logs collected in association with the advertisement selection, and calculates a correlation degree for each of the plurality of collected user behavior logs And transmits the result to the determination unit 22.

At this time, the calculating unit 21 calculates the correlation with the advertisement selection for each of the plurality of user behavior logs collected in association with the advertisement selection using the number of the advertisement selection, which is the number of times the advertisement is selected after each user action log is confirmed Can be calculated.

That is, the calculation unit 21 calculates an average value and a variance value for the number of advertisement selection of each of a plurality of user behavior logs, and calculates an advertisement selection number of each user behavior log using the calculated average value and the variance value And the degree of correlation is calculated through the normalized value (Z).

Further, the calculating unit 21 may calculate a variance value for a value (Z) obtained by normalizing each advertisement selection number for the purpose of reflecting stability, before each user behavior log is determined as matching information for reflecting in the matching algorithm And the final result is calculated as a correlation for each user action log.

The determination unit 22 performs a function of determining matching information.

More specifically, when the determination result of the degree of correlation with the advertisement selection for each of the plurality of collected user log information is received, the determining unit 22 determines at least some of the user behavior logs as a priority having a high degree of correlation with the advertisement selection And determines the selected user action log as matching information for matching the advertisement and the user, and delivers the determination result to the providing unit 23. [

At this time, the determination unit 22 may integrate the first user behavior log collected in the first period and the second user behavior log collected in the second period adjacent to the first period, Selection is made according to the priority with a high degree of correlation with the selection and is determined as matching information for matching the advertisement and the user.

The determination of the matching information by integrating the first user behavior log and the second user behavior log is performed in order to consider that the variance value in the first period and the second period may vary.

Meanwhile, prior to the final determination of the matching information, the determining unit 22 confirms the matching information optimum number through the learned matching algorithm based on the training data corresponding to the user behavior log, The user behavior logs corresponding to the optimum number of matching information items identified in the user behavior log are selected according to the priority degree of correlation and determined as matching information.

Here, when the test data is applied to the matching algorithm instead of the training data, the optimal number of matching information is the number of user behavior logs in the test data corresponding to the point at which the matching rate between the advertisement and the user is derived from the matching algorithm .

The providing unit 23 performs a function of providing matching information.

More specifically, when the matching information determined according to the correlation degree priority is received, the providing unit 23 provides the matching information to the advertisement providing apparatus 30 through the communication unit 2310, To match the received matching information to the matching algorithm to process the matching between the advertisement and the user of the terminal 10 and to provide the terminal 10 with an advertisement matched with the user of the terminal 10 as the processing result.

Hereinafter, the configuration of the advertisement providing apparatus 30 will be described with reference to the configuration of the advertisement providing apparatus 20 according to the embodiment of the present invention.

7 shows a schematic configuration of an advertisement providing apparatus 30 according to an embodiment of the present invention

7, an advertisement providing apparatus 30 according to an embodiment of the present invention includes an acquiring unit 31 for acquiring matching information from the advertisement supporting apparatus 20, a selecting unit 32 for selecting an advertisement, , And a provider 33 for providing an advertisement.

All or at least a part of the configuration of the advertisement providing apparatus 30 including the obtaining unit 31, the selecting unit 32 and the providing unit 33 may be implemented in the form of a software module executed by a processor or a hardware module Or software modules and hardware modules may be implemented in a combined form.

The advertisement providing apparatus 30 according to an embodiment of the present invention further includes a communication unit 34 for supporting communication between the terminal 10 and the advertisement supporting apparatus 20 in addition to the above- .

For reference, the configuration of the communication unit 34 corresponds to the configuration of the communication unit 3310 to be described below with reference to FIG. 8, and a detailed description thereof will be described below.

An advertisement providing apparatus 30 according to an embodiment of the present invention may include an advertisement providing unit 30 for providing an advertisement suitable for a user of the terminal 10 through a core configuration including an obtaining unit 31, a selecting unit 32, . Hereinafter, each configuration in the advertisement providing device 30 for realizing this will be described in detail.

The acquiring unit 31 performs a function of acquiring matching information.

More specifically, when an advertisement request is received from the terminal 10, the acquiring unit 31 transmits matching information to be applied to the matching algorithm to the advertisement supporting apparatus (for example, 20).

In this regard, when the matching information is requested from the advertisement providing apparatus 30, the advertisement supporting apparatus 20 correlates user behavior logs of the optimum number of matching information, which are identified from the matching algorithm, Also determines the selected user activity log as matching information according to the priority and provides the selected user behavior log to the advertisement providing device 30.

In the embodiment of the present invention, the advertisement providing apparatus 20 provides the matching information according to the request of the advertisement providing apparatus 30. However, the present invention is not limited to this, It goes without saying that the advertisement providing apparatus 20 may periodically provide the determined matching information to the advertisement providing apparatus 30 without any separate request.

The selection unit 32 performs a function of selecting an advertisement.

More specifically, when the matching information is received from the advertisement supporting apparatus 20, the selecting unit 32 applies the received matching information to the matching algorithm to process the matching between the advertisement and the user of the terminal 10, The advertisement matching the user of the terminal 10 is selected.

The providing unit 33 performs a function of providing an advertisement.

More specifically, the providing unit 33 provides the selected advertisement to the terminal 10 when the advertisement is selected as a result of applying the matching information received from the advertisement supporting apparatus 20 to the matching algorithm.

In the meantime, each of the configurations of the advertisement providing apparatus 30 described above may be implemented by a software module, a hardware module, or a combination of a software module and a hardware module executed by a processor, in a real hardware system (e.g., a computer system) .

8, a description will be given of a hardware system 3000 implementing each configuration in an advertisement providing apparatus 30 according to an embodiment of the present invention.

On the other hand, among the configurations of the hardware system 3000 shown in Fig. 8, configurations whose names coincide with those in the hardware system 1000 described with reference to Fig. 3 can be regarded as having the same configuration, A specific example or description of the same configuration will be omitted.

8, a hardware system 3000 for implementing an advertisement providing apparatus 30 according to an embodiment of the present invention includes a processor unit 3100, a memory interface unit 3200, (3300). ≪ / RTI >

The processor unit 3100 communicates with the memory unit 3210 through the memory interface unit 3200 to execute various software modules stored in the memory unit 3210 to perform various functions in the hardware system .

Here, the acquiring unit 31, the selecting unit 32, and the providing unit 33, which are components of the advertisement providing apparatus 30 described with reference to FIG. 7, can be stored in the memory unit 3210 in the form of a software module And other operating systems can be additionally stored.

The peripheral device interface unit 3300 serves to enable communication between the processor unit 3100 and the peripheral device.

Here, in the case of a peripheral device, the hardware system 3000 is provided for providing different functions. In an embodiment of the present invention, for example, a communication unit 3310 may be included.

Here, the communication unit 3310 performs a function of providing a communication function with other devices. For example, an antenna system, an RF transceiver, one or more amplifiers, a tuner, one or more oscillators, a digital signal processor, a codec CODEC) chipset, and memory, and the like, and may include known circuits that perform this function.

As a result, each configuration in the advertisement providing apparatus 30 stored in the memory unit 3210 in the form of a software module is stored in the memory interface unit 3100 in the form of a command executed by the processor unit 3100, It is possible to provide the terminal 10 with an advertisement matched with the user of the terminal 10 by processing the interface with the communication unit 3310 via the Internet 3300.

Hereinafter, each configuration of the advertisement providing device 30 will be described in connection with the hardware system 3000 with reference to FIG.

The acquiring unit 31 performs a function of acquiring matching information.

More specifically, when the advertisement request from the terminal 10 is received through the communication unit 3210, the obtaining unit 31 obtains the advertisement for matching with the matching algorithm for the purpose of selecting the advertisement matched with the user of the terminal 10 Requesting the advertisement support apparatus 20 to obtain matching information, and transmitting the acquired matching information to the selection unit 32. [

In this regard, when the matching information is requested from the advertisement providing apparatus 30, the advertisement supporting apparatus 20 correlates user behavior logs of the optimum number of matching information, which are identified from the matching algorithm, Also determines the selected user activity log as matching information according to the priority and provides the selected user behavior log to the advertisement providing device 30.

The selection unit 32 performs a function of selecting an advertisement.

More specifically, when the matching information obtained from the advertisement supporting apparatus 20 is received, the selecting unit 32 applies the received matching information to the matching algorithm to process the matching between the advertisement and the user of the terminal 10, Selects an advertisement matched with the user (10), and delivers the selected advertisement to the provider (33).

The providing unit 33 performs a function of providing an advertisement.

More specifically, the provider 33 provides the received advertisement to the terminal 10 through the communication unit 3310 when the advertisement selected through the matching algorithm applying the matching information is received.

As described above, according to each configuration in the advertisement providing system according to the embodiment of the present invention, matching between advertisements and users in the user behavior log collected in association with the advertisement selection, It is possible to solve the problem of overfitting that can be generated in the matching algorithm based on the existing machine learning by optimally applying the matching information for the optimal number of the features, Thereby improving the advertiser's return on investment (ROI).

Hereinafter, an operation of the advertisement providing system according to the embodiment of the present invention and the operation flow in each configuration in the system will be described.

FIG. 9 shows an operation flow in an advertisement providing system according to an embodiment of the present invention.

First, the advertisement providing apparatus 30 transmits matching information for applying to the matching algorithm for the purpose of selecting an advertisement matched with the user of the terminal 10 when an advertisement request is received from the terminal 10 according to steps S11 and S12, To the advertisement support apparatus 20. [

In response to this, the advertisement supporting apparatus 20 calculates the degree of correlation with the advertisement selection for each of the plurality of user behavior logs collected in association with the advertisement selection in accordance with step S13.

At this time, the advertisement supporting apparatus 20 may calculate the correlation with the advertisement selection for each of the plurality of user action logs collected in association with the advertisement selection using the advertisement selection count, which is the number of times the advertisement is selected after each user action log is confirmed Can be calculated.

That is, the advertisement supporting apparatus 20 calculates an average value and a variance value for the number of advertisement selection times of each of a plurality of user behavior logs, and calculates an advertisement selection number The correlation value is calculated through the normalized value Z.

Furthermore, the advertisement support apparatus 20 may further include a variance value (Z) for a value (Z) obtained by normalizing each advertisement selection number for the purpose of reflecting stability, before each user action log is determined as matching information for reflecting in the matching algorithm Is calculated as a correlation for each user action log.

When the degree of correlation with the advertisement selection for each of the plurality of collected user log information is calculated, at least some of the user behavior logs are correlated with the advertisement selection according to steps S14 and S15 And selects the user action log as the matching information for matching the advertisement and the user.

At this time, the advertisement supporting apparatus 20 may integrate the first user activity log collected in the first period and the second user behavior log collected in the second period adjacent to the first period, And determines the matching information as the matching information for matching the advertisement and the user by selecting according to the priority with a high degree of correlation with the advertisement selection.

The determination of the matching information by integrating the first user behavior log and the second user behavior log is performed in order to consider that the variance value in the first period and the second period may vary.

Meanwhile, prior to finally determining the matching information, the advertisement supporting device 20 checks the optimal number of matching information through the learned matching algorithm based on the training data using the training data corresponding to the user behavior log. The number of user behavior logs corresponding to the optimum number of pieces of matching information identified in the user behavior log is selected according to the correlation degree priority and determined as the matching information.

Here, when the test data is applied to the matching algorithm instead of the training data, the optimal number of matching information is the number of user behavior logs in the test data corresponding to the point at which the matching rate between the advertisement and the user is derived from the matching algorithm .

Then, when the match information is determined according to the correlation degree priority, the advertisement supporting apparatus 20 provides the matching information determined according to step S16 to the advertisement providing apparatus 30. [

Further, when the matching information is received from the advertisement supporting device 20, the advertisement providing device 30 applies the received matching information to the matching algorithm in accordance with step S17 to process the matching between the advertisement and the user of the terminal 10, And selects an advertisement matching with the user of the terminal 10 as a result of the processing.

Thereafter, when the advertisement is selected as a result of applying the matching information received from the advertisement supporting apparatus 20 to the matching algorithm, the advertisement providing apparatus 30 provides the selected advertisement to the terminal 10 according to step S18.

The operation flow in the advertisement providing system according to the embodiment of the present invention is described above, and the operation flow in the advertisement supporting apparatus 20 will be described.

FIG. 10 shows an operation flow in the advertisement supporting apparatus 20 according to an embodiment of the present invention.

First, the calculating unit 21 calculates the degree of correlation with the advertisement selection for each of the plurality of user behavior logs collected in association with the advertisement selection according to steps S21 to S23.

At this time, the calculating unit 21 calculates the correlation with the advertisement selection for each of the plurality of user behavior logs collected in association with the advertisement selection using the number of the advertisement selection, which is the number of times the advertisement is selected after each user action log is confirmed Can be calculated.

That is, the calculation unit 21 calculates an average value and a variance value for the number of advertisement selection of each of a plurality of user behavior logs, and calculates an advertisement selection number of each user behavior log using the calculated average value and the variance value And the degree of correlation is calculated through the normalized value (Z).

Further, the calculating unit 21 may calculate a variance value for a value (Z) obtained by normalizing each advertisement selection number for the purpose of reflecting stability, before each user behavior log is determined as matching information for reflecting in the matching algorithm And the final result is calculated as a correlation for each user action log.

Then, when the degree of correlation with the advertisement selection for each of the plurality of pieces of collected user log information is calculated, the determination unit 22 determines at least some of the user behavior logs according to steps S24 to S26 as a correlation with the advertisement selection And the selected user action log is determined as matching information for matching the advertisement and the user.

At this time, the determination unit 22 may integrate the first user behavior log collected in the first period and the second user behavior log collected in the second period adjacent to the first period, Selection is made according to the priority with a high degree of correlation with the selection and is determined as matching information for matching the advertisement and the user.

The determination of the matching information by integrating the first user behavior log and the second user behavior log is performed in order to consider that the variance value in the first period and the second period may vary.

Meanwhile, prior to the final determination of the matching information, the determination unit 22 confirms the matching information optimum number through the learned matching algorithm based on the training data corresponding to the user behavior log, The user behavior logs corresponding to the optimum number of matching information items identified in the user behavior log are selected according to the priority degree of correlation and determined as matching information.

Here, when the test data is applied to the matching algorithm instead of the training data, the optimal number of matching information is the number of user behavior logs in the test data corresponding to the point at which the matching rate between the advertisement and the user is derived from the matching algorithm .

Then, when the matching information determined according to the priority degree of correlation is determined, the providing unit 23 provides the matching information determined in step S27 to the advertisement providing apparatus 30 through the communication unit 2310, 30 to match the received matching information to the matching algorithm to process the matching between the advertisement and the user of the terminal 10 and to provide the terminal 10 with an advertisement matching the user of the terminal 10 as the processing result do.

Hereinafter, the operation flow in the advertisement providing apparatus 30 will be described after the description of the operation flow in the advertisement supporting apparatus 20 according to the embodiment of the present invention.

11 shows an operation flow in the advertisement providing apparatus 30 according to an embodiment of the present invention.

First, in a case where an advertisement request is received from the terminal 10 according to steps S31 and S32, the obtaining unit 31 obtains matching information to be applied to the matching algorithm for the purpose of selecting an advertisement matched with the user of the terminal 10 And obtains it by requesting the advertisement supporting apparatus 20.

In this regard, when the matching information is requested from the advertisement providing apparatus 30, the advertisement supporting apparatus 20 correlates user behavior logs of the optimum number of matching information, which are identified from the matching algorithm, Also determines the selected user activity log as matching information according to the priority and provides the selected user behavior log to the advertisement providing device 30.

Then, when the matching information is received from the advertisement supporting apparatus 20, the selecting unit 32 applies the received matching information to the matching algorithm according to steps S33 and S34 to process the matching between the advertisement and the user of the terminal 10 , And selects an advertisement that matches the user of the terminal 10 as a result of the processing.

Thereafter, the providing unit 33 provides the selected advertisement to the terminal 10 according to step S35 when the advertisement is selected as a result of applying the matching information received from the advertisement supporting apparatus 20 to the matching algorithm.

As described above, according to the operation flow in each configuration of the advertisement providing system and the system according to the embodiment of the present invention, regarding the machine learning based matching algorithm, among the user action logs collected in association with the advertisement selection Since the problem of overfitting that may occur in the existing machine learning based matching algorithm can be solved by applying the optimal number of matching information for matching between the advertisement and the user, , And can further improve the advertiser's return on investment (ROI).

It should be understood that the functional operations and subject matter implementations described herein may be implemented as digital electronic circuitry, or may be embodied in computer software, firmware, or hardware, including the structures disclosed herein, and structural equivalents thereof, . Implementations of the subject matter described herein may be implemented as one or more computer program products, i. E. One or more modules relating to computer program instructions encoded on a type of program storage medium for execution by, or control of, the operation of the processing system Can be implemented.

The computer-readable medium can be a machine-readable storage device, a machine-readable storage substrate, a memory device, a composition of matter that affects the machine readable propagation type signal, or a combination of one or more of the foregoing.

The term "system" or "device" as used herein encompasses all apparatuses, apparatus, and machines for processing data, including, for example, a programmable processor unit, a computer or a multiprocessor unit or a computer. The processing system may include, in addition to the hardware, code that forms an execution environment for the computer program upon request, such as, for example, code comprising the processor portion firmware, a protocol stack, a database management system, an operating system, have.

A computer program (also known as a program, software, software application, script or code) may be written in any form of programming language, including compiled or interpreted language, a priori or procedural language, Components, subroutines, or other units suitable for use in a computer environment. A computer program does not necessarily correspond to a file in the file system. The program may be stored in a single file provided to the requested program, or in multiple interactive files (e.g., a file storing one or more modules, subprograms, or portions of code) (E.g., one or more scripts stored in a markup language document). A computer program may be deployed to run on multiple computers or on one computer, located on a single site or distributed across multiple sites and interconnected by a communications network.

On the other hand, computer readable media suitable for storing computer program instructions and data include semiconductor memory devices such as, for example, EPROM, EEPROM and flash memory devices, such as magnetic disks such as internal hard disks or external disks, Non-volatile memory, media and memory devices, including ROM and DVD-ROM disks. The processor portion and the memory may be supplemented by, or integrated into, a special purpose logic circuit.

Implementations of the subject matter described herein may include, for example, a back-end component such as a data server, or may include a middleware component, such as an application server, or may be a web browser or a graphical user, for example a user, who may interact with an implementation of the subject- Front-end components such as client computers with interfaces, or any combination of one or more of such back-end, middleware, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication, such as, for example, a communications network.

While the specification contains a number of specific implementation details, it should be understood that they are not to be construed as limitations on the scope of any invention or claim, but rather on the description of features that may be specific to a particular embodiment of a particular invention Should be understood. Likewise, the specific features described herein in the context of separate embodiments may be implemented in combination in a single embodiment. Conversely, various features described in the context of a single embodiment may also be implemented in multiple embodiments, either individually or in any suitable subcombination. Further, although the features may operate in a particular combination and may be initially described as so claimed, one or more features from the claimed combination may in some cases be excluded from the combination, Or a variant of a subcombination.

It is also to be understood that although the present invention is described herein with particular sequence of operations in the drawings, it is to be understood that it is to be understood that it is to be understood that all such illustrated acts have to be performed or that such acts must be performed in their particular order or sequential order, Can not be done. In certain cases, multitasking and parallel processing may be advantageous. Also, the separation of the various system components of the above-described embodiments should not be understood as requiring such separation in all embodiments, and the described program components and systems will generally be integrated together into a single software product or packaged into multiple software products It should be understood that

As such, the present specification is not intended to limit the invention to the specific terminology presented. Thus, while the present invention has been described in detail with reference to the above examples, those skilled in the art will be able to make adaptations, modifications, and variations on these examples without departing from the scope of the present invention. The scope of the present invention is defined by the appended claims rather than the detailed description and all changes or modifications derived from the meaning and scope of the claims and their equivalents are to be construed as being included within the scope of the present invention do.

According to an advertisement supporting apparatus, an operation method thereof, and an advertisement providing apparatus of the present invention, in relation to a machine learning based matching algorithm for matching between an advertisement and a user, Since the matching information for matching can be selectively applied, it is not only the use of the related technology but also the possibility of commercialization or sales of the applied device, as it exceeds the limit of the existing technology. This is an invention that is industrially applicable because it can be done.

10: Terminal
11: App Advertisement Section 12: Web Advertising Section
20: Ad-supported device
21: calculating section 22: determining section
23: Offering
30: Ad serving device
31: Acquiring unit 32:
33: Offering

Claims (12)

A processor for performing operations associated with matching an advertisement to a user; And
Wherein at least one instruction executed via the processor comprises a stored memory,
Wherein the at least one instruction comprises:
A calculation instruction for calculating a degree of correlation with the advertisement selection based on the number of times of selection of the advertisement, which is the number of times the advertisement is selected after each user behavior log is confirmed, for each of the plurality of user behavior logs collected in association with the advertisement selection; And
Selecting at least a part of the user behavior logs among the plurality of user behavior logs according to a priority having a high degree of correlation with the advertisement selection and determining the selected user behavior log as matching information for matching the advertisement and the user And an instruction to display the advertisement. The method according to claim 1,
The degree of correlation with the advertisement selection,
Wherein the average of the number of advertisement selection times of each of the plurality of user behavior logs is calculated using a value obtained by normalizing the number of advertisement selection times of each of the plurality of user behavior logs,
Wherein the user action log is calculated to have a higher value than the remaining user action log when the number of times the advertisement is selected is greater than the average value. The method according to claim 1,
The plurality of user behavior logs comprising:
At least one of a first user behavior log collected during a first period and a second user behavior log collected during a second period neighboring the first period,
The determination command includes:
And selects at least a part of the user behavior logs having a high degree of correlation between the first user behavior log and the second user behavior log as the matching information. The method according to claim 1,
The determination command includes:
Wherein the matching information determining unit determines the matching information optimum number by using the learned matching algorithm by using the training data including the user behavior log and determines the user behavior log for the matching information optimum number as the matching information. . 5. The method of claim 4,
The optimal number of matching information,
And the number of user action logs in the test data corresponding to a time point at which a matching rate between an advertisement and a user is derived from the matching algorithm is highest when test data is applied to the matching algorithm instead of the training data. Device. A processor for performing an operation related to providing an advertisement; And
Wherein at least one instruction executed via the processor comprises a stored memory,
Wherein the at least one instruction comprises:
Acquiring at least a part of a plurality of user action logs collected in association with advertisement selection of a specific user as matching information for matching an advertisement with the specific user; And
And a selection command for selecting an advertisement matched with the specific user based on the matching information and providing the selected advertisement to the specific user. The method according to claim 6,
The matching information may include:
At least some user behavior logs selected in accordance with a priority having a high correlation with the advertisement selection among the plurality of user behavior logs,
The degree of correlation with the advertisement selection,
And the number of times the advertisement is selected after each user action log is confirmed. Calculating a degree of correlation with the advertisement selection based on the number of times the advertisement is selected after each user action log is confirmed for each of the plurality of user action logs collected in association with the advertisement selection; And
Determining at least a part of the user behavior logs among the plurality of user behavior logs according to a priority having a high degree of correlation with the advertisement selection and determining the selected user behavior log as matching information for matching an advertisement and a user The method comprising the steps of: 9. The method of claim 8,
The degree of correlation with the advertisement selection,
Wherein the average of the number of advertisement selection times of each of the plurality of user behavior logs is calculated using a value obtained by normalizing the number of advertisement selection times of each of the plurality of user behavior logs,
Wherein the user action log is calculated to have a value higher than the remaining user action log when the number of advertisement selection times is larger than the average value. 9. The method of claim 8,
The plurality of user behavior logs comprising:
At least one of a first user behavior log collected during a first period and a second user behavior log collected during a second period neighboring the first period,
Wherein,
Wherein at least a part of the user behavior logs having a high degree of correlation with the advertisement selection from the first user behavior log and the second user behavior log is selected and determined as the matching information. 9. The method of claim 8,
Wherein,
Wherein the matching information determining unit determines the matching information optimum number by using the learned matching algorithm by using the training data including the user behavior log and determines the user behavior log for the matching information optimum number as the matching information. Lt; / RTI > 12. The method of claim 11,
The optimal number of matching information,
And the number of user action logs in the test data corresponding to a time point at which a matching rate between an advertisement and a user is derived from the matching algorithm is highest when test data is applied to the matching algorithm instead of the training data. Method of operation of the device.

Images