Method, system, and computer program for expanding and utilizing data using kana combination

Abstract

To provide a method, a system, and a computer program for expanding and utilizing data using pseudonym association.SOLUTION: A method for expanding and utilizing data using pseudonym association includes the steps of: receiving pseudonym association data in which first-party information of a customer company and third-party information of a company other than the customer company are pseudonymized and combined; generating an interpretation model for data estimation by modeling the pseudonym association data; estimating the third-party information using target information of the customer company through the interpretation model; and providing a service related to the customer company with the estimated third-party information.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The following description relates to a technology for utilizing data of other companies.

Background Art

[0002] Companies set advertising exposure targets or provide personalized customized advertisements based on an understanding of their own customers.

[0003] As an example of a technology for providing targeted advertisements, Patent Document 1 (publication date: January 22, 2008) discloses a technology for selecting preferred advertisements using a user profile.

[0004] Most companies do not have sufficient information about their own customers, so they are deepening their understanding of customers while using information of other companies.

[0005] As methods for securing information of other companies, there are methods such as linking and utilizing own company data (data of one company) and data of three companies using third-party cookies, and methods for utilizing data (data of two companies) provided with the consent of customers.

[0006] However, due to the strengthening of personal information protection in recent years, it has become difficult to utilize third-party data using cookies, and it has also begun to be difficult to secure customer data.

Prior Art Documents

Patent Documents

[0007]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0008] Protect the privacy of individuals through pseudonymization while enabling the continuous and legal use of data from other companies.

[0009] Build a model that estimates data from other companies from the company's own data using pseudonymized combined data, so as to be able to estimate the latest information.

[0010] Estimate data from other companies using the model built with pseudonymized combined data, so as to be able to expand the data to be utilized.

[0011] By combining not only the company's own data but also the data from other companies estimated by the model based on the pseudonymization infrastructure, it can be utilized for enhancing targeting such as advertising execution and personalized recommendations.

Means for Solving the Problem

[0012] A method for providing data from other companies executed by a computer system, wherein the computer system includes at least one processor configured to execute computer-readable instructions included in a memory, and the method for providing data from other companies includes: receiving, by the at least one processor, pseudonymized combined data obtained by pseudonymizing and combining the information of the customer company itself and the information of other companies that are not the customer company; generating, by the at least one processor, an interpretation model for data estimation by modeling the pseudonymized combined data; estimating, by the at least one processor, information of other companies using the target information of the customer company by means of the interpretation model; and providing, by the at least one processor, the estimated information of other companies for a service related to the customer company.

[0013] According to one aspect, the generating step may include a step of anonymizing and de-identifying the pseudonymized combined data.

[0014] According to another aspect, the generating step may include anonymizing the pseudonymized data by at least one of a process of smoothing the pseudonymized data and a process of adding an error to the pseudonymized data.

[0015] According to still another aspect, the generating step may generate the interpretation model by a method of summarizing the pseudonymized data in a table form, a method of summarizing the pseudonymized data using the similarity between items, or a method of modeling the pseudonymized data using a logistic regression model or a deep learning model.

[0016] According to still another aspect, the generating step may include deleting the pseudonymized data after the generation of the analysis model is completed.

[0017] According to still another aspect, the providing step may include providing, to the customer company, extended data in a form in which the target information and the estimated other company information are concatenated.

[0018] According to still another aspect, the providing step may include using the target information and the estimated other company information for advertising targeting for the customer company.

[0019] According to still another aspect, in the estimating step, the data of the other company is estimated from the own data of the target customers of the customer company using the interpretation model, and in the providing step, the extended data obtained by concatenating the own data and the data of the other company may be defined as an advertising targeting option for the target customers.

[0020] According to still another aspect, the receiving step may receive pseudonymized data in a form in which the pseudonymization key has been removed from an external joining specialist organization.

[0021] According to still another aspect, the receiving step may receive the pseudonym combined data obtained by combining the pseudonym data of the customer company and the pseudonym data of at least one other company selected as a target company for information utilization by the customer company.

[0022] Provided is a computer program recorded on a computer-readable recording medium for causing a computer to execute the other company data providing method.

[0023] Provided is a computer system including at least one processor configured to execute computer-readable instructions included in a memory, the at least one processor including a process of receiving pseudonym combined data obtained by pseudonymizing and combining the information of the customer company itself and the information of other companies that are not the customer company, a process of generating an interpretation model for data estimation by modeling the pseudonym combined data, a process of estimating other company information using the target information of the customer company by the interpretation model, and a process of providing the estimated other company information for a service related to the customer company.

Advantages of the Invention

[0024] According to an embodiment of the present invention, while protecting the privacy of individuals by pseudonym combination, other company data can be continuously and legally estimated and utilized.

[0025] According to an embodiment of the present invention, by constructing a model for estimating other company data from its own data using pseudonym combined data, the accuracy of the model can be ensured and the currency of the data can be guaranteed.

[0026] According to an embodiment of the present invention, by estimating other company data using the connection of data by a model constructed with pseudonym combined data, the data to be utilized can be expanded.

[0027] According to an embodiment of the present invention, not only company's own data but also combined with data of other companies estimated by a pseudonymized model can be utilized to enhance targeting such as advertisement execution and personalized recommendation.

Brief Description of Drawings

[0028]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Embodiments for Carrying Out the Invention

[0029] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

[0030] Embodiments of the present invention relate to a technique for utilizing data of other companies.

[0031] Embodiments including the matters specifically disclosed in this specification can provide a platform that can continuously and legally estimate and utilize data of other companies while protecting the privacy of individuals by kana binding.

[0032] The other-company data providing system according to an embodiment of the present invention may be realized by at least one computer system, and the other-company data providing method according to an embodiment of the present invention may be executed by at least one computer system included in the other-company data providing system. At this time, in the computer system, a computer program according to an embodiment of the present invention may be installed and executed, and the computer system may execute the other-company data providing method according to an embodiment of the present invention according to the control of the executed computer program. The above-described computer program may be recorded on a computer-readable recording medium in order to be combined with the computer system and cause the computer to execute the other-company data providing method.

[0033] FIG. 1 is a diagram showing an example of a network environment in an embodiment of the present invention. The network environment of FIG. 1 shows an example including a plurality of electronic devices 110, 120, 130, 140, a plurality of servers 150, 160, and a network 170. Such FIG. 1 is merely an example for explaining the invention, and the number of electronic devices and the number of servers are not limited as in FIG. 1. Further, the network environment of FIG. 1 is merely an example of the environments applicable to the present embodiment, and the environments applicable to the present embodiment are not limited to the network environment of FIG. 1.

[0034] The plurality of electronic devices 110, 120, 130, 140 may be fixed terminals or mobile terminals realized by a computer device. Examples of the plurality of electronic devices 110, 120, 130, 140 include smartphones, mobile phones, navigation devices, PCs (personal computers), notebook PCs, digital broadcast terminals, PDAs (Personal Digital Assistants), PMPs (Portable Multimedia Players), tablets, and the like. As an example, in FIG. 1, a smartphone is shown as an example of the electronic device 110. However, in the embodiments of the present invention, the electronic device 110 may mean any one of various physical computer devices that can communicate with other electronic devices 120, 130, 140 and / or servers 150, 160 via the network 170 using substantially wireless or wired communication methods.

[0035] The communication method is not limited, and it may include not only a communication method using a communication network that the network 170 can include (for example, a mobile communication network, a wired Internet, a wireless Internet, a broadcast network), but also short-range wireless communication between devices. For example, the network 170 may include any one or more of networks such as a PAN (personal area network), a LAN (local area network), a CAN (campus area network), a MAN (metropolitan area network), a WAN (wide area network), a BBN (broadband network), and the Internet. Further, the network 170 may include any one or more of network topologies including a bus network, a star network, a ring network, a mesh network, a star-bus network, a tree or hierarchical network, etc., but is not limited thereto.

[0036] Each of servers 150 and 160 may be implemented by one or more computer devices that communicate with a plurality of electronic devices 110, 120, 130, and 140 via a network 170 to provide instructions, code, files, content, services, and the like. For example, server 150 may be a system that provides a first service to a plurality of electronic devices 110, 120, 130, and 140 connected via network 170, and server 160 may also be a system that provides a second service to a plurality of electronic devices 110, 120, 130, and 140 connected via network 170. As a more specific example, server 150 may provide, as the first service, a service (such as a customer data platform service, for example) targeted by an application, which is a computer program installed and executed on a plurality of electronic devices 110, 120, 130, and 140, to the plurality of electronic devices 110, 120, 130, and 140 through the application. As another example, server 160 may provide, as the second service, a service that distributes files for installation and execution of the above-described application to a plurality of electronic devices 110, 120, 130, and 140.

[0037] FIG. 2 is a block diagram showing an example of a computer system according to an embodiment of the present invention. Each of the plurality of electronic devices 110, 120, 130, and 140 and each of servers 150 and 160 described above may be implemented by the computer system 200 shown in FIG. 2.

[0038] Such a computer system 200 may include, as shown in FIG. 2, a memory 210, a processor 220, a communication interface 230, and an input / output interface 240.

[0039] Memory 210 is a computer-readable recording medium and may include a random access memory (RAM), a read-only memory (ROM), and a permanent mass storage device such as a disk drive. Here, a permanent mass storage device such as a ROM or a disk drive may be included in computer system 200 as a separate permanent recording device distinct from memory 210. Also, an operating system and at least one program code may be recorded in memory 210. Such software components may be loaded into memory 210 from a computer-readable recording medium other than memory 210. Such another computer-readable recording medium may include computer-readable recording media such as a floppy (registered trademark) drive, a disk, a tape, a DVD / CD-ROM drive, and a memory card. In other embodiments, the software components may be loaded into memory 210 through a communication interface 230 that is not a computer-readable recording medium. For example, the software components may be loaded into memory 210 of computer system 200 based on a computer program installed by a file received via network 170.

[0040] Processor 220 may be configured to process instructions of a computer program by performing basic arithmetic, logic, and input / output operations. The instructions may be provided to processor 220 by memory 210 or communication interface 230. For example, processor 220 may be configured to execute instructions received according to a program code recorded in a recording device such as memory 210.

[0041] The communication interface 230 may provide a function for the computer system 200 to communicate with other devices (for example, the recording device described above) via the network 170. As an example, requests, instructions, data, files, etc. generated by the processor 220 of the computer system 200 according to program codes recorded in a recording device such as the memory 210 may be transmitted to other devices via the network 170 under the control of the communication interface 230. Conversely, signals, instructions, data, files, etc. from other devices may be received by the computer system 200 through the communication interface 230 of the computer system 200 via the network 170. Signals, instructions, data, etc. received through the communication interface 230 may be transmitted to the processor 220 or the memory 210, and files, etc. may be recorded in a recording medium (the permanent recording device described above) that the computer system 200 can further include.

[0042] The input / output interface 240 may be means for interfacing with the input / output device 250. For example, the input device may include devices such as a microphone, a keyboard, or a mouse, and the output device may include devices such as a display or a speaker. As another example, the input / output interface 240 may be means for interfacing with a device in which functions for input and output are integrated into one, such as a touch screen. The input / output device 250 may be composed of one device with the computer system 200.

[0043] Also, in other embodiments, the computer system 200 may include fewer or more components than the components in FIG. 2. However, it is not necessary to clearly show most of the components of the prior art in the figure. For example, the computer system 200 may be realized to include at least a part of the input / output device 250 described above, or may further include other components such as a transceiver or a database.

[0044] Hereinafter, specific embodiments of a method and a system for expanding and utilizing data using pseudonym binding will be described.

[0045] In order for a company to utilize the information of other companies, it obtains consent from customers and then receives the information, or utilizes methods such as matching data of different companies by using third-party cookies or data similarity.

[0046] This embodiment relates to a method of constructing an interpretation model with legal data using pseudonym binding, and a technique for continuously estimating and utilizing data of different companies.

[0047] The computer system 200 according to this embodiment may provide a customer data platform service for a client by connecting to a dedicated application installed on the client or a web / mobile site related to the computer system 200.

[0048] Such a processor 220 may control the computer system 200 to execute the steps included in the method for providing data of other companies described below. For example, the processor 220 and the components of the processor 220 may be realized to execute the code of the operating system included in the memory 210 and the instructions of at least one program.

[0049] The processor 220 may read the necessary instructions from the memory 210 in which the instructions related to the control of the computer system 200 are loaded. In this case, the read instructions may include instructions for controlling the processor 220 to execute the method for providing data of other companies described below.

[0050] The steps included in the method for providing data of other companies described below may be executed in an order different from that shown in the figures, or some of the steps may be omitted, or additional processes may be further included.

[0051] The steps included in the method of providing data of other companies may be executed on the server 150. However, depending on the embodiment, at least a part of the steps may also be executed on the client or another server (for example, the server 160, etc.).

[0052] FIG. 3 is a flowchart showing an example of a method that can be executed by a computer system according to an embodiment of the present invention.

[0053] The server 150 realized by the computer system 200 according to this embodiment provides a customer data platform that can be used for advertising, marketing, etc. in a form that links data of customer companies that use the service (that is, the customer information of the customer companies themselves) and data of other companies other than the customer companies (for example, the customer information of the server 150, the customer information of partner companies that can be linked with the server 150, etc.) as a service form targeted at companies.

[0054] Referring to FIG. 3, at step 310, the processor 220 may generate an interpretation model for data estimation by modeling the anonymized combined data obtained by anonymizing and combining the information of the customer company itself and the information of other companies other than the customer company. As an example, after a customer company selects at least one target company to be utilized from the customer data platform on the server 150, it may request the construction of an interpretation model. Thereby, the processor 220 may request an external combining specialized agency (an external combining specialized agency designated by the government) to combine the anonymized information for the data provided by anonymizing the customer company and the target company selected by the customer company. The processor 220 may receive the anonymized combined data in which the anonymized data of the customer company and the anonymized data of the target company are combined from the combining specialized agency. The processor 220 may use the anonymized combined data to construct, as an interpretation model for data estimation, a model that estimates the data of other companies (target companies) with the customer company data as input, a model that estimates the customer company data with the data of other companies as input, and the like. As a method of modeling the anonymized combined data between the customer company and the target company, a method of summarizing the anonymized combined data in a table format, a method of summarizing the anonymized combined data using the similarity between items, a method of summarizing the anonymized combined data using a logistic regression model, a method of summarizing the anonymized combined data using a deep learning model, and the like may be utilized. After constructing the analysis model, the processor 220 may delete the anonymized combined data received from the combining specialized agency.

[0055] In step 320, when the processor 220 is given target information from the customer company, it may use the analysis model constructed in step 310 to estimate other company information using the target information of the customer company. That is, the processor 220 may use the target information of the customer company as the input to the analysis model and predict other company information corresponding to the target information of the customer company by the analysis model. For example, when it is assumed that an interpretation model is constructed from the pseudonym-linked data combining the product purchase history of the customer company's own customers of Company A and the search history of the customer company's own customers of Company B (the target company), the search history of Company B may be estimated from the product purchase history of Company A for the customer company's own customers.

[0056] In step 330, the processor 220 may provide extended data that concatenates the target information of the customer company and the other company information estimated as target information by the interpretation model for the customer company. The processor 220 may provide extended data in a form that concatenates the customer company data and the other company data estimated as customer company data for all of the customer company's own customers or for a group of the customer company's own customers selected by the customer company.

[0057] The processor 220 may utilize the extended data of the customer for services targeted at the customer. As an example, the processor 220 may utilize the extended data of the customer company for advertising targeting, personalized recommendations, marketing, etc. related to the customer company. When the customer company intends to provide an advertisement to its own customers, it may interpret the target customers of the customer company as other company data estimated by the interpretation model and define an advertising targeting option. For example, in an environment where the server 150 providing the search service cooperates with an advertising platform, the interpretation model may be used to estimate the search history from the target customer history of the customer company, and the extended data including the estimated search history may be defined as an advertising targeting option. At this time, the advertising platform may execute an advertisement for the target customers of the customer company according to the targeting option defined by the data extended by the estimation of other company information.

[0058] FIG. 4 is an exemplary diagram for explaining the process of anonymizing the data of each company in one embodiment of the present invention.

[0059] FIG. 4 shows the anonymized data of Company A and the anonymized data of Company B.

[0060] Each company itemizes the data of its own customers according to the purpose of utilization. For example, the anonymized data of Company A may include customer segment information for each customer of Company A, and the anonymized data of Company B may include advertising targeting information for each customer of Company B.

[0061] Each company constructs customer data at the level of user identifiers (such as IDs, cookies, etc.) commonly held by both companies. At this time, when constructing anonymized data, in order to anonymize user identification information, a combined anonymization key may be generated using user identification information and a salt or hash, etc. The user information may be anonymized so that a specific individual cannot be identified without using additional information.

[0062] FIGS. 5 to 6 are exemplary diagrams for explaining anonymized combined data obtained by combining the anonymized data of each company in one embodiment of the present invention.

[0063] FIG. 5 shows an example of anonymized combination of the anonymized data of Company A and the anonymized data of Company B in FIG. 4. The combination of anonymized information is performed by an external combination specialized agency designated by the government, and the server 150 may receive the anonymized combined data in which the anonymized data of Company A and the anonymized data of Company B are combined from the combination specialized agency.

[0064] FIG. 5 shows the result of anonymized combination of the anonymized data of Company A and the anonymized data of Company B in the form of an inner join. The combination specialized agency may combine the data of both companies for customers (anonymized combination key) commonly held by Company A and Company B.

[0065] When the data of the data utilization agency is used for anonymized combination, anonymized combination in the form of a left join with the data of the data utilization agency is also possible.

[0066] Referring to FIG. 6, in the combination specialized institution, when the combination of the pseudonym data of Company A and the pseudonym data of Company B is completed, the pseudonym combination key may be deleted from the combined data and provided to the server 150. That is, the server 150 may receive and utilize the pseudonym combined data in a form in which the pseudonym combination key is deleted.

[0067] Hereinafter, the process of constructing an analysis model using the pseudonym combined data will be described.

[0068] FIG. 7 is a diagram showing an example of a problem to be handled in constructing an analysis model according to an embodiment of the present invention.

[0069] In order to construct an analysis model, assuming that the pseudonym combined data obtained by combining the pseudonym data of Company A and the pseudonym data of Company B is as shown in FIG. 7, the problem of whether an arbitrary item user group of Company A can be analogized as the item user information of Company B is handled.

[0070] If the items of Company A are A1,..., A n and the items of Company B are B1,..., B m when the user U i may be defined as [A i1 ,..., A in , B i1 ,..., B im .

[0071] As an example, the processor 220 may construct an analysis model by summarizing the pseudonym combined data in a table format.

[0072] As the simplest method of summarizing the original without loss of information,

Number

[0073] A method of merging and processing data that can be identified as one person with other groups may be used. While maintaining the homogeneity within each group to the maximum extent, the distance between groups progresses in a direction of moving apart. In order to minimize the loss of data information due to merging, a distance reflecting the importance of each item may be used, or a method of anonymizing some items with low importance may be adopted.

[0074] As another example, the processor 220 may construct an interpretation model by a method of summarizing pseudonymized data using the similarity between items.

[0075] The relationship matrix between the items of Company A and the items of Company B can be processed and utilized in combination with the user information of Company B to infer the user information of Company A. According to Equation (1), since S is the correspondence information between items and does not have user identifier information, it can be said that it is relatively safe data. However, in reality, since B -1 does not exist, it is also a method that cannot be used immediately.

[0076]

Number

[0077] This can be transformed into the form of item-based collaborative filtering, which is one of the simple recommendation systems. It is a method of assigning high scores to items similar to the items previously consumed by the user. Similarity can be applied in various ways, such as cosine similarity and Jaccard similarity. Modifications such as using KNN (k-nearest neighbor) are also possible.

[0078] S is a similarity matrix of the items of Company A and Company B, and may be defined as in Equation (2).

[0079]

Number

[0080] The consumption score of a user for the items of Company A may be defined as in Equation (3).

[0081]

Number

[0082] As another example, the processor 220 may construct an interpretation model by a method of summarizing the kana combination data using a logistic regression model.

[0083] When assuming that the joint distribution is the product of the marginal distributions, the distribution for all combinations can be easily obtained as follows.

[0084] Logistic regression model for individual items:

Number

[0085] The joint distribution for all combinations of items may be defined as in Equation (4).

[0086]

Number

[0087] By expanding this concept, the conditions of two items are regarded as one event

Number

[0088] As another example, the processor 220 may construct an analysis model by a method of summarizing kana-concatenated data using a deep learning model.

[0089] FIG. 8 is a diagram showing an example of the structure of an encoder-decoder using an RNN (recurrent neural network) in one embodiment of the present invention.

[0090] The Seq2Seq (sequence-to-sequence) method, which is a model that outputs a sequence of another domain from the input sequence, is used. Referring to FIG. 8, when the input sequence (x1, x2, x3) and the output sequence (y1, y2) are configured as items of Company A and items of Company B, respectively, it will play a role of translating (interpreting) into items of Company B when the items of Company A are input.

[0091] However, it should not be limited to such a modeling method, and any modeling method capable of summarizing kana-concatenated data is applicable.

[0092] The modeling of the matching method that only uses the similarity of cookies and data has low accuracy. On the contrary, by constructing an interpretation model using the combination of pseudonym information, the accuracy of the model can be ensured by the method that most accurately combines in terms of the Personal Information Protection Law.

[0093] Furthermore, the processor 220 may perform de-identification processing on the results of the analysis model. As an example, in the process of constructing the interpretation model, the processor 220 may anonymize the pseudonymized data to eliminate the possibility of re-identifying the model results in advance. As another example, the processor 220 may perform de-identification processing by means of a smoothing process or adding an error in the modeling process so that individuals cannot be identified from the estimation results of the analysis model even when using the pseudonymized data without modification.

[0094] Also, the processor 220 may guarantee anonymity by verifying the possibility of re-identification. For example, the processor 220 may apply k-anonymity to check whether there are k or more rows with the same attributes in the results estimated using all the data held by the company as input.

[0095] FIGS. 9 to 10 are exemplary diagrams for explaining the data expansion process using the analysis model in an embodiment of the present invention.

[0096] Referring to FIG. 9, the processor 220 may estimate the information (N i ) of Company B by using the target information (A i ) of Company A with the analysis model 90 modeled by the pseudonymized data of Company A and Company B.

[0097] Although the data of all companies is not fixed and continues to change, by using the analysis model 90, the estimation results of the analysis model 90 also change with the fluctuations of the input data held by the company, so the latest information can be estimated. Also, by using the analysis model 90 constructed with pseudonymized data without actually holding the pseudonymized data, continuously integrated data can be utilized.

[0098] The processor 220 uses the information (A i ) of Company A, which is the customer company, to obtain the information (N i) can be estimated to expand Company A's data. As shown in FIG. 10, the expanded data of Company A may be configured in a form that concatenates Company A's data for all of Company A's customers and the information of Company B estimated by the analysis model 90.

[0099] When Company A intends to execute an advertisement targeting Company A's VIP customers, in the expanded data of FIG. 10, if A1 is an item of Company A's VIP status (1 if VIP, 0 if not VIP), and N i is an item of Company B's advertising targeting option, the characteristics of users for whom A1 is 1 (Company A's VIP group) may be represented by N i . By executing an advertisement based on such a combination of characteristics (a combination of targeting options), it is possible to execute an advertisement targeting Company A's VIP group that is intended to be targeted.

[0100] Assuming that Company B is an advertising platform company, since there is no information regarding Company A's VIP customers on Company B's advertising platform, it is impossible to execute an advertisement targeting Company A's VIPs, and only targeting with low resolution is possible as the characteristics of customers analyzed using Company A's data. For example, if the main age of VIPs is in their 30s and they are female, an advertisement targeted at the broad conditions of "in their 30s & female" will be executed.

[0101] On the other hand, according to this embodiment, by using both Company A's data and the data of Company B estimated as Company A's data on Company B's advertising platform, the characteristics of actual Company A's VIP customers can be represented by Company B's data, and it becomes possible to reach Company A's customers through more accurate and precise targeting. In this embodiment, it is possible to estimate Company B's history using Company A's data of "in their 30s & female", and thereby define more specific targeting options.

[0102] Thus, according to the embodiments of the present invention, while protecting an individual's privacy through pseudonym binding, it is possible to continuously and legally estimate and utilize data of other companies. Also, according to the embodiments of the present invention, by constructing a model for estimating data of other companies from its own data using pseudonym-bound data, the accuracy of the model can be ensured and the currency of the data can be guaranteed. Further, according to the embodiments of the present invention, by estimating data of other companies using the connection between data by a model constructed with pseudonym-bound data, the data to be utilized can be expanded. Furthermore, according to the embodiments of the present invention, not only its own data but also by combining with data of other companies estimated by a pseudonym-binding-based model, it can be utilized for enhancing targeting such as advertisement execution and personalized recommendation.

[0103] The apparatus described above may be implemented by hardware components, software components, and / or a combination of hardware components and software components. For example, the apparatus and components described in the embodiments may be implemented using one or more general-purpose computers or special-purpose computers, such as a processor, a controller, an ALU (arithmetic logic unit), a digital signal processor, a microcomputer, an FPGA (field programmable gate array), a PLU (programmable logic unit), a microprocessor, or various devices capable of executing instructions and responding. The processing device may execute an operating system (OS) and one or more software applications executed on the OS. Also, the processing device may access data, record, manipulate, process, and generate data in response to the execution of the software. For the sake of convenience of understanding, although it may be described that one processing device is used, those skilled in the art will understand that the processing device may include a plurality of processing elements and / or a plurality of types of processing elements. For example, the processing device may include a plurality of processors or one processor and one controller. Also, other processing configurations such as parallel processors are possible.

[0104] Software may include a computer program, code, instructions, or a combination of one or more of these, and may configure a processing device to operate as desired or command the processing device independently or collectively. Software and / or data may be embodied in any kind of machine, component, physical device, virtual equipment, computer recording medium or device for the purpose of being interpreted by a processing device or providing instructions or data to the processing device. Software may be distributed on a computer system connected by a network and recorded or executed in a distributed state. Software and data may be recorded on one or more computer-readable recording media.

[0105] The method according to an embodiment may be realized in the form of program instructions executable by various computer means and recorded on a computer-readable medium. At this time, the medium may be one that continuously records a program executable by a computer, or one that temporarily records it for execution or download. Further, the medium may be various recording means or storage means in a form in which a single or a plurality of hardwares are combined, and may be not only a medium directly connected to a certain computer system, but also one that is distributed and exists on a network. Examples of the medium include magnetic media such as hard disks, floppy (registered trademark) disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, magneto-optical media such as floptical disks, and those configured to record program instructions such as ROMs, RAMs, and flash memories. Also, as examples of other media, there are included recording media and storage media managed by an app store that distributes an application, and sites, servers, etc. that supply and distribute other various softwares.

[0106] As described above, the embodiments have been described based on limited embodiments and drawings. However, those skilled in the art will be able to make various modifications and variations from the above description. For example, even if the described technology is executed in an order different from the described method, and / or the components such as the described system, structure, device, circuit, etc. are combined or assembled in a form different from the described method, or opposed or replaced by other components or equivalents, appropriate results can be achieved.

[0107] Therefore, even if they are different embodiments, as long as they are equivalent to the scope of the claims, they belong to the scope of the appended claims.

Description of Reference Numerals

[0108] 110, 120, 130, 140: Electronic devices 150, 160: Servers 170: Network

Claims

1. A method for providing data of other companies executed on a computer system, comprising: The computer system includes at least one processor configured to execute computer-readable instructions included in a memory; The method for providing data of other companies includes: Receiving, by the at least one processor, the company's own information of the customer company and the anonymized combined data obtained by anonymizing and combining the information of other companies other than the customer company; Generating, by the at least one processor, an analysis model for data estimation by modeling the anonymized combined data; Estimating, by the at least one processor, other company information using the target information of the customers of the customer company by the analysis model, and Providing, by the at least one processor, the estimated other company information for services related to the customer company. Including, The anonymization process generates a first anonymized combination key by using a sort or hash on the user identifier of the customer included in the own information of the customer company, and generates a second anonymized combination key by using a sort or hash on the user identifier of the customer included in the information of other companies other than the customer company. The combining process combines the item information related to the customer included in the own information and the item information related to the customer included in the other company information when the first anonymized combination key and the second anonymized combination key match. Modeling the anonymized combined data includes identifying a matrix of items related to the own information of the customer company and a matrix of items related to the information of other companies other than the customer company in the anonymized combined data based on the own information of the customer company. The target information of the customer is information for identifying the users included in the own information for which the customer company intends to execute an advertisement. The estimating step includes the at least one processor estimating the search history included in the other company information from the purchase history of the customer input in the analysis model. A method for providing data of other companies.

2. The generating step includes: Anonymizing the anonymized combined data so that customers cannot be identified. The anonymization process includes applying a sort or hash to the user identifier of the customer. The method for providing data of other companies according to Claim 1.

3. The generating step is a step of making it impossible to identify a customer in the pseudonymized combined data by at least one of a process of smoothing a pseudonymized combined key generated by using a sort or a hash on a user identifier of the customer in the pseudonymized combined data, and a process of adding an error to the pseudonymized combined key in the pseudonymized combined data The method for providing data of another company according to claim 1, comprising the above.

4. The generating step is generating an interpretation model by a method of constructing the pseudonymized combined data in a table format, or a method of constructing the pseudonymized combined data by using a similarity between items, or a method of modeling the pseudonymized combined data by using a logistic regression model or a deep learning model The method for providing data of another company according to any one of claims 1 to 3, characterized by the above.

5. The generating step is a step of deleting the pseudonymized combined data after completion of the generation of the analysis model The method for providing data of another company according to any one of claims 1 to 3, comprising the above.

6. The providing step is a step of providing extended data in a form in which the target information and the estimated data of another company are concatenated to the customer The method for providing data of another company according to any one of claims 1 to 3, comprising the above.

7. The providing step is a step of using the target information and the estimated data of another company for advertisement targeting for the customer company The method for providing data of another company according to any one of claims 1 to 3, comprising the above.

8. The estimating step is estimating a search history of data of another company from a purchase history of in-house data of a target customer of the customer company by using the analysis model The providing step is a step of executing an advertisement for the target customer based on the in-house data and the data extended by the estimation of the data of another company The method for providing data of another company according to any one of claims 1 to 3, comprising the above.

9. The receiving step is a step of receiving pseudonymized combined data in a form in which a pseudonymized combined key generated by using a sort or a hash on a user identifier of a customer is deleted from an external combination specialized agency The method for providing data of another company according to any one of claims 1 to 3, characterized by including the above.

10. A computer program for causing a computer device to execute a method for providing data of other companies according to any one of Claims 1 to 3.

11. A computer system comprising: at least one processor configured to execute computer-readable instructions included in a memory and wherein the at least one processor receives anonymized combined data obtained by anonymizing and combining the customer company's own information and the information of other companies other than the customer company; generates an analysis model for data estimation by modeling the anonymized combined data; estimates other company information using the target information of the customer company by the analysis model; and provides the estimated other company information for services related to the customer company. The at least one processor processes wherein the anonymization process generates a first anonymized combination key using a sort or hash on the customer identifiers of the customers included in the own information of the customer company, and generates a second anonymized combination key using a sort or hash on the customer identifiers of the customers included in the other company information other than the customer company, and the combining process combines the item information related to the customer included in the own information and the item information related to the customer included in the other company information when the first anonymized combination key and the second anonymized combination key match; modeling the anonymized combined data includes identifying a matrix of items related to the own information of the customer company and a matrix of items related to the other company information other than the customer company in the anonymized combined data based on the own information of the customer company; the target information of the customer is information for identifying a user included in the own information for which the customer company intends to execute an advertisement, and the estimating step includes the at least one processor estimating a search history included in the other company information from the purchase history of the customer input in the analysis model. A computer system.

12. The at least one processor characterized in that it anonymizes the anonymized combined data so that customers cannot be identified, wherein the anonymization includes applying a sort or hash to the customer identifier. The computer system according to claim 11.

13. The at least one processor The process of smoothing the anonymized coupling key generated by using a salt or a hash for the customer's user identifier in the anonymized coupling data, and making it impossible to identify the customer in the anonymized coupling data by at least one of the processes of adding an error to the anonymized coupling key in the anonymized coupling data The computer system according to claim 11, characterized in that

14. The at least one processor Generating an interpretation model by a method of constructing the anonymized coupling data in a table format, or a method of constructing the anonymized coupling data using the similarity between items, or a method of modeling the anonymized coupling data using a logistic regression model or a deep learning model The computer system according to any one of claims 11 to 13, characterized in that

15. The at least one processor After the generation of the analysis model is completed, deleting the anonymized coupling data The computer system according to any one of claims 11 to 13, characterized in that

16. The at least one processor Providing extended data in a form in which the target information and the estimated information of other companies are linked for the customers of the customer company The computer system according to any one of claims 11 to 13, characterized in that

17. The at least one processor Using the target information and the estimated information of other companies for advertising targeting for the customer company The computer system according to any one of claims 11 to 13, characterized in that

18. The at least one processor Using the analysis model to estimate the search history of other company data from the purchase history of the target customer's own company data of the customer company, Executing an advertisement for the target customer based on the data extended by the estimation of the own company data and the other company data The computer system according to any one of claims 11 to 13, characterized in that

Images