Method, device, and program for providing personalized recommendations through collection and analysis of zero party data

Abstract

A method for providing customized recommendations through the collection and analysis of Zero Party Data according to various embodiments of the present invention is disclosed. The method may include the steps of: creating a user interface comprising a plurality of queries and an answer input window for inputting an answer to each query and providing it to a user terminal; collecting Zero Party Data (ZPD) based on a response to the user interface; acquiring First Party Data (FPD) corresponding to the user terminal; and generating customized recommendation information through integrated analysis based on the Zero Party Data and the First Party Data.

Description

Technology Field

[0001] The present invention relates to a method for providing customized recommendation information, and more specifically, to a method for providing personalized recommendation information by analyzing zero-party data. Background Technology

[0003] Data-driven personalized recommendation systems are widely utilized in fields such as e-commerce, digital marketing, content recommendation, customer relationship management (CRM), healthcare, finance, and education, and are evolving into technologies that provide optimized information by analyzing users' interests and preferences. Such personalized recommendation technology is establishing itself as a key element in enhancing user experience and boosting the competitiveness of corporate services and products.

[0004] Conventional personalized recommendation technologies have primarily operated by analyzing and predicting user preferences using first-party data (user past behavior data) and third-party data (data acquired from third parties). First-party data refers to data generated directly by users during their use of websites or applications, including, for example, click records, search history, purchase history, viewing time, and app usage patterns. In contrast, third-party data is collected from platforms providing external data or third-party advertising networks and has been utilized for more extensive user profile analysis.

[0005] However, the utilization of third-party data is becoming increasingly difficult due to the recent strengthening of personal information protection and data regulations. With the implementation of various data protection laws, such as the European Union's General Data Protection Regulation (GDPR) and the U.S. California Consumer Privacy Act (CCPA), the collection of data based on third-party cookies is being restricted, and consequently, several issues faced by conventional personalized recommendation systems are being highlighted.

[0006] Due to these environmental changes, a new approach to data collection methods is required. Conventional personalized recommendation systems primarily predicted preferences based on users' past behavioral patterns; however, this method alone has limitations in that it is difficult to reflect the user's clear intentions. In particular, it is difficult to provide appropriate recommendations to new users (the "Cold Start" problem) who lack sufficient behavioral data, and it is often difficult to accurately predict user preferences using behavioral data alone. For instance, even if a user clicks on a specific product multiple times, it is difficult to determine whether this is merely interest or an actual purchase intention, and there is a risk of misinterpreting temporary interest as long-term interest.

[0007] Furthermore, existing recommendation systems suffer from a problem in that they fail to sufficiently verify the reliability of the data. If recommendations are based solely on user click data or search history, unintended clicks or search patterns can influence the results, and there is a possibility that recommendations will be provided that do not align with the user's specific interests. This can lead to reduced recommendation accuracy and a degraded user experience.

[0008] To address these issues, the importance of Zero Party Data, which is provided directly by users, has recently been increasing. Zero Party Data refers to data explicitly provided by users, including information directly entered by them regarding their interests, preferences, requirements, and purchase intentions. For example, this includes information such as categories preferred by the user, interest in specific brands, and desired personalized service settings.

[0009] Zero-party data, being information directly entered by users, can provide more accurate and reliable data. When combined with existing behavioral data (First Party Data), it enables more precise personalized recommendations. Furthermore, the process of encouraging users to voluntarily provide data increases transparency in data collection and helps secure user trust.

[0010] However, conventional recommendation systems suffered from low data acquisition rates due to a lack of technology for effectively collecting and analyzing zero-party data and inadequate interfaces that induce users to voluntarily provide data. Furthermore, the insufficient verification process for the accuracy of user-provided data meant there was a possibility that recommendation quality could deteriorate if inaccurate responses were reflected.

[0011] To address these issues, a new method is required to effectively collect zero-party data and organically integrate it with first-party data for analysis. This enables more precise personalized recommendations, enhances data reliability, and is expected to improve the user experience while complying with privacy regulations. Prior art literature

[0012] Republic of Korea Published Patent Application No. 10-2010-0011609 (February 3, 2010) The problem to be solved

[0013] The problem that the present invention aims to solve is devised in response to the aforementioned background technology, and is to provide highly reliable personalized recommendation information by effectively collecting users' zero-party data and integrating and analyzing it with first-party data.

[0014] The problems that the present invention aims to solve are not limited to those mentioned above, and other unmentioned problems will be clearly understood by a person skilled in the art from the description below. means of solving the problem

[0016] A method for providing customized recommendations through the collection and analysis of Zero Party Data according to an embodiment of the present invention for solving the aforementioned problem is disclosed. The method may include the steps of: creating a user interface including a plurality of queries and an answer input window for inputting answers to each query and providing it to a user terminal; collecting Zero Party Data (ZPD) based on the response to the user interface; acquiring First Party Data (FPD) corresponding to the user terminal; and generating customized recommendation information through integrated analysis based on the Zero Party Data and the First Party Data.

[0017] In an alternative embodiment, the user interface includes a selection input screen for obtaining a selectable response from a user of the user terminal, including the plurality of queries and the answer input window, and a user input screen for obtaining a direct response from the user. The step of creating and providing the user interface to the user terminal includes: determining a plurality of queries related to the user's interests based on the user's previous responses and behavioral pattern information; selecting at least two response items from a group of response candidates based on the user's real-time context information, the previous responses, and the behavioral pattern information; and creating a selection input screen including the determined plurality of queries and response items corresponding to each query. The behavioral pattern information may be information that serves as a basis for analyzing actual interests and behavioral trends through comparison with the user's selected responses.

[0018] In an alternative embodiment, the step of creating and providing the user interface to the user terminal comprises receiving real-time environment information from the user terminal, determining which screen to provide among the selection input screen and the user input screen based on the real-time environment information, and providing the user interface including the determined screen to the user terminal, wherein the real-time environment information is information regarding the situation in which a response input occurs and the state of the user terminal, and may include at least one of device type information, network status information, location information, and response speed information.

[0019] In an alternative embodiment, the method may include the steps of: analyzing the response pattern of the zero-party data to calculate consistency information with the previous response of the same user; comparing the first-party data and the zero-party data to calculate reliability information of the response; if the reliability score is less than a preset threshold, providing an additional user interface including a re-inquiry to the user terminal and performing correction on the response based on the response to the additional user interface; and generating the zero-party data by applying weights to the response according to the consistency information.

[0020] In an alternative embodiment, the method may include the steps of determining an incentive based on completeness, reliability information and participation frequency corresponding to the response, providing the determined incentive to the user terminal, and dynamically adjusting the timing of the provision of the user interface based on the time at which the response is collected and the participation pattern.

[0021] In an alternative embodiment, the step of collecting zero-party data based on a response to the user interface may include: refining the response according to a predefined data structure to generate zero-party data; encrypting the zero-party data using a pre-set encryption algorithm; and storing the encrypted zero-party data in a database and restricting data retrieval and utilization according to an access control policy.

[0022] In an alternative embodiment, the step of generating customized recommendation information through an integrated analysis based on the zero-party data and the first-party data may include: constructing a training dataset using zero-party data and first-party data obtained from a plurality of users; training a machine learning-based integrated analysis model for user segment allocation using the training dataset; processing the zero-party data and the first-party data as input to the integrated analysis model and allocating them to user segments; generating user profile information based on the allocated user segments and the zero-party data; and generating customized recommendation information based on the user profile information.

[0023] In an alternative embodiment, the step of generating customized recommendation information through an integrated analysis based on the zero-party data and the first-party data may include: acquiring additional zero-party data corresponding to the user; generating change information by comparing the additional zero-party data with the user profile information; updating the user profile information based on the generated change information; and evaluating the result of updating the user profile information to determine whether to change the segment to which the user belongs.

[0024] According to another embodiment of the present invention, an apparatus for providing customized recommendations through the collection and analysis of Zero Party Data is disclosed. The apparatus includes a memory for storing one or more instructions and a processor for executing one or more instructions stored in the memory, and the processor can perform the aforementioned method for providing customized recommendations through the collection and analysis of Zero Party Data by executing one or more instructions.

[0025] According to another embodiment of the present invention, a computer program stored on a computer-readable recording medium is disclosed. The computer program is combined with a computer, which is hardware, to perform a method for providing customized recommendations through the collection and analysis of Zero Party Data.

[0026] Other specific details of the present invention are included in the detailed description and drawings. Effects of the invention

[0028] According to one embodiment of the present invention, user zero-party data can be effectively collected and integrated with first-party data for analysis to provide more precise customized recommendation information. This enables highly reliable personalized recommendations that reflect the user's explicit preferences and can complement the limitations of existing behavior-based recommendation systems.

[0029] Furthermore, by evaluating the reliability of collected zero-party data and ensuring data consistency through machine learning-based analysis, the impact of inaccurate responses on recommendation results can be minimized. Consequently, more accurate and optimized user profiles can be constructed, and the quality of personalized recommendations can be improved.

[0030] Additionally, by continuously updating user profiles to reflect real-time user behavior data and recent changes in interests, and reassigning user segments as needed, a customized recommendation system can be implemented that flexibly responds to changing user needs. This enables the provision of an optimized recommendation experience for both new and existing users, thereby improving long-term user satisfaction.

[0031] The effects of the present invention are not limited to those mentioned above, and other unmentioned effects will be clearly understood by a person skilled in the art from the description below. Brief explanation of the drawing

[0033] FIG. 1 is an exemplary diagram schematically illustrating a system for implementing a method for providing customized recommendations through the collection and analysis of Zero Party Data related to an embodiment of the present invention. FIG. 2 is a hardware configuration diagram of a computing device that performs a method for providing customized recommendations through the collection and analysis of Zero Party Data related to an embodiment of the present invention. FIG. 3 illustrates a flowchart exemplarily showing a method for providing customized recommendations through the collection and analysis of Zero Party Data related to an embodiment of the present invention. FIG. 4 illustrates a flowchart that exemplarily shows the process of creating a user interface related to one embodiment of the present invention. FIG. 5 is a flowchart exemplarily illustrating the process of collecting zero-party data related to an embodiment of the present invention. FIG. 6 is a flowchart exemplarily illustrating the process of generating user profile information by integrating zero-party data and first-party data related to an embodiment of the present invention. Specific details for implementing the invention

[0034] Various embodiments are now described with reference to the drawings. In this specification, various descriptions are provided to facilitate an understanding of the invention. However, it is evident that these embodiments can be practiced without such specific descriptions.

[0035] As used herein, terms such as “component,” “module,” “system,” etc. refer to computer-related entities, hardware, firmware, software, combinations of software and hardware, or executions of software. For example, a component may be a process, processor, object, execution thread, program, and / or computer running on a processor, but is not limited thereto. For example, both an application running on a computing device and the computing device itself may be a component. One or more components may reside within a processor and / or execution thread. A component may be localized within a single computer. A component may be distributed among two or more computers. Additionally, these components may be executed from various computer-readable media having various data structures stored therein. Components may communicate through local and / or remote processes, for example, according to signals having one or more data packets (e.g., data from a component interacting with another component in a local system or distributed system, and / or data transmitted through signals to other systems and networks such as the Internet).

[0036] Furthermore, the term "or" is intended to mean an implicit "or" rather than an exclusive "or." That is, unless otherwise specified or evident from the context, "X uses A or B" is intended to mean one of the natural implicit substitutions. In other words, if X uses A; if X uses B; or if X uses both A and B, "X uses A or B" may apply to any of these cases. Additionally, the term "and / or" as used herein should be understood to refer to and include all possible combinations of one or more of the enumerated related items.

[0037] Additionally, the terms “comprising” and / or “comprising” should be understood to mean that such features and / or components are present. However, the terms “comprising” and / or “comprising” should be understood not to exclude the presence or addition of one or more other features, components and / or groups thereof. Furthermore, unless otherwise specified or clearly evident from the context to indicate a singular form, the singular in this specification and claims should generally be interpreted to mean “one or more.”

[0038] Those skilled in the art should recognize that the various exemplary logical blocks, configurations, modules, circuits, means, logics, and algorithmic steps described in connection with the embodiments disclosed herein may be implemented in electronic hardware, computer software, or a combination of both. To clearly exemplify the interchangeability of hardware and software, various exemplary components, blocks, configurations, means, logics, modules, circuits, and steps have been generally described above in terms of their functionality. Whether such functionality is implemented in hardware or software depends on the specific application and design constraints imposed on the overall system. Skilled technicians may implement the described functionality in various ways for each specific application. However, such decisions regarding implementation should not be construed as moving out of the scope of the invention.

[0039] The description of the presented embodiments is provided to enable those skilled in the art to use or practice the present invention. Various modifications to these embodiments will be apparent to those skilled in the art. The general principles defined herein may be applied to other embodiments without departing from the scope of the present invention. Thus, the present invention is not limited to the embodiments presented herein. The present invention should be interpreted in the broadest possible scope consistent with the principles and novel features presented herein.

[0040] In this specification, the term "computer" refers to any type of hardware device comprising at least one processor, and may be understood to include software configurations operating on said hardware device according to the embodiments. For example, the term "computer" may be understood to include smartphones, tablet PCs, desktops, laptops, and user clients and applications running on each of these devices, but is not limited thereto.

[0041] Hereinafter, embodiments of the present invention will be described in detail with reference to the attached drawings.

[0042] Each step described in this specification is described as being performed by a computer, but the subject of each step is not limited thereto, and depending on the embodiment, at least some of each step may be performed on different devices.

[0044] FIG. 1 is an exemplary diagram schematically illustrating a system for implementing a method for providing customized recommendations through the collection and analysis of Zero Party Data related to an embodiment of the present invention.

[0045] As illustrated in FIG. 1, a system according to embodiments of the present invention may include a computing device (100), a user terminal (200), an external server (300), and a network (400). The components illustrated in FIG. 1 are exemplary, and additional components may exist or some of the components illustrated in FIG. 1 may be omitted. The computing device (100), the external server (300), and the user terminal (200) according to embodiments of the present invention may mutually transmit and receive data for a system according to embodiments of the present invention through a network (400).

[0046] A network (400) according to embodiments of the present invention can use various wired communication systems such as a Public Switched Telephone Network (PSTN), xDSL (x Digital Subscriber Line), RADSL (Rate Adaptive DSL), MDSL (Multi Rate DSL), VDSL (Very High Speed ​​DSL), UADSL (Universal Asymmetric DSL), HDSL (High Bit Rate DSL), and a Local Area Network (LAN).

[0047] In addition, the network (400) presented here may use various wireless communication systems such as CDMA (Code Division Multi Access), TDMA (Time Division Multi Access), FDMA (Frequency Division Multi Access), OFDMA (Orthogonal Frequency Division Multi Access), SC-FDMA (Single Carrier-FDMA), and other systems.

[0048] A network (400) according to embodiments of the present invention may be configured regardless of the mode of communication, such as wired or wireless, and may be configured as various communication networks, such as a Personal Area Network (PAN) or a Wide Area Network (WAN). Additionally, the network (400) may be a known World Wide Web (WWW) and may utilize wireless transmission technologies used for short-range communication, such as Infrared Data Association (IrDA) or Bluetooth. The technologies described in this specification may be used in other networks as well as the networks mentioned above.

[0049] A computing device (100) (hereinafter referred to as 'computing device (100)') that performs a method for providing personalized recommendations through the collection and analysis of Zero Party Data related to one embodiment of the present invention can perform the function of inducing a user to voluntarily provide information, verifying collected data, and providing personalized recommendation information by integrating and analyzing it with First Party Data.

[0050] Specifically, the computing device (100) provides a user interface for acquiring zero-party data in conjunction with a user terminal, evaluates reliability based on the collected zero-party data, and can generate and update user profile information by reflecting the analysis results. Here, user profile information may refer to a personalized data structure that includes the user's explicit preferences (e.g., Zero Party Data) and behavior-based information (e.g., First Party Data) and reflects the user's interests, preferred products or services, past response patterns, consumption tendencies, and usage history.

[0051] According to the embodiment, user profile information is continuously updated and can be optimized to reflect the user's latest interests and behavioral patterns whenever new zero-party data is collected.

[0052] In an embodiment, the computing device (100) can continuously optimize personalized custom recommendation information by assigning and adjusting user segments by reflecting the user's latest behavior data and response data in real time.

[0053] Based on this configuration, the data scarcity issue inherent in conventional behavior-based recommendation methods can be resolved, and more precise personalized recommendation information can be provided by reflecting users' explicit preferences. In particular, it improves the accuracy of initial recommendations for new users and enables real-time response to changes in user preferences, thereby enhancing the adaptability of the recommendation system.

[0054] In addition, in the embodiments, by utilizing machine learning based on collected Zero Party Data (ZPD) and First Party Data (FPD) to automatically update user profiles and reassign user segments as needed, optimized recommendations become possible not only for individual users but also for groups sharing specific interests. This encourages active user participation and can improve long-term user satisfaction.

[0055] A detailed explanation of the method for providing customized recommendations through the collection and analysis of Zero Party Data provided by the computing device (100) and the effects caused by it will be described in detail later with reference to FIG. 3.

[0056] In the embodiment, only one computing device (100) in FIG. 1 is illustrated, but it will be obvious to those skilled in the art that more computing devices may also be included within the scope of the invention and that the computing device (100) may include additional components. That is, the computing device (100) may be composed of multiple computing devices. In other words, a set of multiple nodes may constitute the computing device (100).

[0057] According to one embodiment of the present invention, the computing device (100) may be a server that provides cloud computing services. More specifically, the computing device (100) may be a server that provides cloud computing services, which are a type of internet-based computing, where information is processed by another computer connected to the internet rather than the user's computer. Cloud computing services may be services that store data on the internet and allow users to access necessary data or programs anytime and anywhere via internet access without installing them on their own computers, and allow data stored on the internet to be easily shared and transmitted through simple operations and clicks. Furthermore, cloud computing services may not only simply store data on a server on the internet but also allow users to perform desired tasks using the functions of applications provided on the web without installing separate programs, and may be services that allow multiple people to work while sharing documents simultaneously. Additionally, cloud computing services may be implemented in at least one form among IaaS (Infrastructure as a Service), PaaS (Platform as a Service), SaaS (Software as a Service), a virtual machine-based cloud server, and a container-based cloud server. That is, the computing device (100) of the present invention may be implemented in at least one form among the aforementioned cloud computing services. The specific description of the aforementioned cloud computing service is merely an example and may include any platform for establishing the cloud computing environment of the present invention.

[0058] A user terminal (200) according to an embodiment of the present invention may refer to any type of node(s) within a system having a mechanism capable of communicating with a computing device (100). The user terminal (200) is a terminal capable of providing Zero Party Data or receiving customized recommendation information through information exchange with the computing device (100), and may include a device possessed or used by the user.

[0059] For example, the user terminal (200) may be a terminal associated with a user inputting Zero Party Data, and the user can use their terminal to check the query and response items provided by the computing device (100) and input their preferences and interests. Additionally, the user can receive personalized recommendation information provided by the computing device (100) and search for products or services based on said information.

[0060] Additionally, the user terminal (200) may be a terminal that receives personalized recommendation information that is updated in real time based on the user's behavioral data (First Party Data). The user can check how their profile is updated through interaction with the computing device (100), and provide additional responses or adjust their preferences as needed.

[0061] In an additional embodiment, the user terminal (200) may be a terminal used by an administrator or a service provider. For example, a data manager may monitor the analysis results of zero-party data and first-party data collected from the computing device (100) and manage the data for system optimization. Additionally, it is possible to analyze the data change trends of a specific user group (segment) through the administrator terminal and adjust the method of providing personalized content.

[0062] In an embodiment, the user terminal (200) may include a display and perform the function of receiving user input and providing customized information. The user terminal (200) may include a network interface for connection with a computing device (100) and may be implemented to enable data transmission and reception in a wired / wireless environment.

[0063] The user terminal (200) may include various entity(s) within a system having a mechanism for communicating with the computing device (100). For example, the user terminal (200) may include a PC (Personal Computer), a notebook, a mobile terminal, a smartphone, a tablet PC, a wearable device, etc., and may include all types of terminals capable of connecting to a wired / wireless network.

[0064] Additionally, the user terminal (200) may be implemented in a software environment such as an API (Application Programming Interface) and a plug-in, and may be configured to transmit and receive data in conjunction with a client application or application source.

[0065] In one embodiment, an external server (300) may be connected to a computing device (100) via a network (400), and may provide various information and data necessary for the computing device (100) to perform a method of providing customized recommendations through the collection and analysis of Zero Party Data, or receive, store, and manage result data derived from the performance of said method.

[0066] For example, the external server (300) may be a storage server separately provided outside the computing device (100) and may support the storage, management, and analysis of large-scale user data, including zero-party data and first-party data. Additionally, it may perform the role of storing and managing training data, validation data, and test data for training a machine learning-based integrated analysis model and a customized recommendation model for user segment allocation.

[0067] In various embodiments, information stored on an external server (300) can be used as a dataset for training the machine learning-based integrated analysis model of the present invention. This data is an essential element for optimizing the creation and updating of user profile information, and as new zero-party data and first-party data are continuously collected, it supports the model in performing learning in real time and optimizing performance.

[0068] Training data stored on an external server (300) is used to learn the relationship between the user's explicit preferences (Zero Party Data) and behavior-based data (First Party Data), thereby enabling the generation of more precise user profile information. Additionally, validation data serves to evaluate and adjust the performance of the trained model, and test data can be used to evaluate the recommendation accuracy for new user data.

[0069] The external server (300) can be implemented in various forms and can contribute to further improving the performance of the method for providing customized recommendations through the collection and analysis of zero-party data by efficiently exchanging data with the computing device (100). In particular, it performs a data management function to continuously update user profile and segment information, and supports the computing device (100) in providing optimized recommendation information by analyzing user data in real time.

[0070] That is, the external server (300) may be a server that performs the role of storing and managing a data set for training the machine learning-based user segment allocation and recommendation model of the present invention. The external server (300) may be implemented as a digital device, such as a laptop computer, notebook computer, desktop computer, cloud storage system, distributed data server, and other computing devices capable of network connectivity. Additionally, it may be configured in various forms, such as an API server for providing web services, a database server for big data analysis, and a cloud server that supports real-time updates of the recommendation model, and the present invention is not limited thereto.

[0072] FIG. 2 is a hardware configuration diagram of a computing device that performs a method for providing customized recommendations through the collection and analysis of Zero Party Data related to an embodiment of the present invention.

[0073] Referring to FIG. 2, a computing device (100) that performs a method for providing customized recommendations through the collection and analysis of Zero Party Data according to one embodiment of the present invention may include one or more processors (110), a memory (120) that loads a computer program (151) executed by the processor (110), a bus (130), a communication interface (140), and a storage (150) that stores the computer program (151). Here, FIG. 2 only illustrates components related to the embodiment of the present invention. Therefore, a person skilled in the art to which the present invention pertains will understand that other general-purpose components may be included in addition to the components illustrated in FIG. 2.

[0074] According to one embodiment of the present invention, the processor (110) can typically handle the overall operation of the computing device (100). The processor (110) can provide or process appropriate information or functions to a user or user terminal by processing signals, data, information, etc. that are input or output through the components described above, or by running an application program stored in memory (120).

[0075] Additionally, the processor (110) can perform operations for at least one application or program for executing the method according to embodiments of the present invention, and the computing device (100) may have one or more processors.

[0076] According to one embodiment of the present invention, the processor (110) may be composed of one or more cores and may include a processor for data analysis and deep learning, such as a central processing unit (CPU), a general purpose graphics processing unit (GPGPU), and a tensor processing unit (TPU) of a computing device.

[0077] The processor (110) can read a computer program stored in memory (120) and perform a method for providing customized recommendations through the collection and analysis of Zero Party Data according to one embodiment of the present invention.

[0078] In various embodiments, the processor (110) may further include RAM (Random Access Memory) and ROM (Read-Only Memory) for temporarily and / or permanently storing signals (or data) processed within the processor (110). Additionally, the processor (110) may be implemented in the form of a system-on-chip (SoC) comprising at least one of a graphics processing unit, RAM, and ROM.

[0079] Memory (120) stores various data, instructions and / or information. Memory (120) may load a computer program (151) from storage (150) to execute a method / operation according to various embodiments of the present invention. When the computer program (151) is loaded into memory (120), the processor (110) may perform the method / operation by executing one or more instructions constituting the computer program (151). Memory (120) may be implemented as a volatile memory such as RAM, but the technical scope of the present disclosure is not limited thereto.

[0080] The bus (130) provides communication functions between components of the computing device (100). The bus (130) can be implemented as various types of buses, such as an address bus, a data bus, and a control bus.

[0081] The communication interface (140) supports wired and wireless internet communication of the computing device (100). Additionally, the communication interface (140) may support various communication methods other than internet communication. To this end, the communication interface (140) may be configured to include a communication module well known in the art of the present invention. In some embodiments, the communication interface (140) may be omitted.

[0082] Storage (150) can store a computer program (151) non-temporarily. When performing a process of providing customized recommendations through the collection and analysis of Zero Party Data via a computing device (100), storage (150) can store various information necessary to provide the process of providing customized recommendations through the collection and analysis of Zero Party Data.

[0083] The storage (150) may be configured to include non-volatile memory such as ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), flash memory, a hard disk, a removable disk, or any form of computer-readable recording medium well known in the art to which the present invention belongs.

[0084] A computer program (151) may include one or more instructions that cause a processor (110) to perform a method / operation according to various embodiments of the present invention when loaded into memory (120). That is, the processor (110) may perform the method / operation according to various embodiments of the present invention by executing the one or more instructions.

[0085] In one embodiment, a computer program (151) may include one or more instructions for performing a method of providing customized recommendations through the collection and analysis of Zero Party Data, the method comprising: creating a user interface including a plurality of queries and an answer input window for inputting an answer to each query and providing it to a user terminal; collecting zero party data based on a response to the user interface; acquiring first party data corresponding to the user terminal; and generating customized recommendation information through integrated analysis based on zero party data and first party data.

[0086] The steps of the method or algorithm described in connection with embodiments of the present invention may be implemented directly in hardware, implemented as a software module executed by hardware, or implemented by a combination thereof. The software module may reside in RAM (Random Access Memory), ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), Flash Memory, a hard disk, a removable disk, a CD-ROM, or any form of computer-readable recording medium well known in the art to which the present invention belongs.

[0087] The components of the present invention may be implemented as a program (or application) and stored on a medium to be executed in combination with a computer, which is hardware. The components of the present invention may be implemented as software programming or software elements, and similarly, embodiments may be implemented in programming or scripting languages ​​such as C, C++, Java, assembler, etc., including various algorithms implemented as combinations of data structures, processes, routines, or other programming configurations. Functional aspects may be implemented as algorithms executed on one or more processors.

[0089] FIG. 3 illustrates a flowchart exemplarily showing a method for providing customized recommendations through the collection and analysis of Zero Party Data related to an embodiment of the present invention.

[0090] The steps illustrated in FIG. 3 may be changed in order as needed, and at least one step may be omitted or added. The steps of FIG. 3 are merely one embodiment of the present invention, and the scope of the present invention is not limited thereto.

[0091] According to one embodiment of the present invention, a method for providing customized recommendations through the collection and analysis of Zero Party Data may include the step (S100) of creating a user interface including a plurality of queries and an answer input window for inputting an answer to each query, and providing it to a user terminal (200).

[0092] The user interface of the present invention can be configured to incorporate gamification elements to induce voluntary user participation and secure a higher response rate. Users can naturally express their preferences through an intuitive and fun selection process, rather than a simple question-based response method, thereby improving the quality and reliability of Zero Party Data. For example, users can express their seasonal preference through a simple A / B choice interface where they select between "Summer" and "Winter," and the immersion of the interface can be enhanced by providing immediate feedback after the selection. As another example, during the process of selecting a preferred travel destination between "Mountains" and "Sea," additional customized questions may be dynamically provided based on the selection result. The specific descriptions regarding the aforementioned choices are merely examples, and the present invention is not limited thereto.

[0093] In this way, by applying game elements, response fatigue can be reduced and continuous data collection can be enabled, thereby providing the effect of improving the accuracy of personalized recommendation information.

[0094] In one embodiment, the user interface may include a selection input screen for quickly identifying user preferences and a user input screen for collecting more free responses. Specifically, the user interface may include a selection input screen for obtaining a selectable response from a user of a user terminal and a user input screen for obtaining a direct response from the user, including a plurality of question and answer input windows.

[0095] According to an embodiment, the selection input screen may be configured to provide a query including a predefined set of response candidates so that the user can express a clear preference.

[0096] For example, this could involve having the user select either "indoor activities" or "outdoor activities" in response to the question, "Which activity do you prefer on the weekend?" If the user selects "outdoor activities," the corresponding response data can be utilized as zero-party data reflecting the user's lifestyle preferences.

[0097] As another example, in response to the question "What do you consider more important when choosing a beverage?", two response options—"taste" or "health"—can be provided. If the user selects "health," that response can be used as foundational data to determine whether the user prefers health-oriented consumption patterns.

[0098] In this way, the selection input screen is configured to allow users to respond intuitively, enabling the rapid and efficient collection of zero-party data.

[0099] In addition, in the embodiment, the user input screen may be configured in a way that allows the user to respond more freely, and may support text input, voice input, or other forms of direct input methods.

[0100] For example, if a query such as "Please write down your favorite summer activity" is provided on a user input screen, the user can freely enter their preferences, such as "swimming" or "camping." Through this, zero-party data reflecting the user's specific opinions or additional information can be obtained, in addition to the data provided by structured multiple-choice responses.

[0101] As such, the user interface of the present invention is configured to include both a selection input screen and a user input screen, thereby enabling the rapid identification of the user's clear preferences while simultaneously collecting more in-depth opinions.

[0102] According to the embodiment, utilizing a selection input screen facilitates the collection of structured data, thereby enabling the effective acquisition of Zero Party Data (ZPD) through the induction of rapid and intuitive responses. On the other hand, the user input screen allows for a more detailed reflection of the user's specific experiences or preferences, offering the advantage of collecting data that is difficult to capture using a simple binary selection method.

[0103] In addition, by combining two input methods, data collection tailored to the user's individual response style is possible, thereby enabling the derivation of richer insights. For example, some users may prefer quick and simple multiple-choice responses, while certain users may wish to input their own opinions in detail. By reflecting these differences and flexibly responding to the user's method of providing data, the interface of the present invention can provide the effect of increasing the response rate and improving the quality of the collected data.

[0104] In other words, the combination of selection input screens and user input screens contributes to securing both the quantity and quality of data simultaneously, and based on this, it is possible to build more sophisticated user profiles and improve the accuracy of personalized recommendations.

[0105] According to one embodiment, a computing device (100) can dynamically generate a user interface to be provided to a user terminal (200) in order to effectively identify the user's preferences and provide personalized recommendations. The computing device (100) can dynamically configure a set of question and response candidates in a manner that allows the user to respond most intuitively by comprehensively considering the user's previous responses, behavioral pattern information, and real-time context information, and can generate a user interface that reflects this.

[0106] More specifically, referring to FIG. 4, the step of creating and providing a user interface to a user terminal may include the step (S110) of determining a plurality of queries related to the user's interests based on the user's previous response and behavior pattern information.

[0107] In the embodiments, behavioral pattern information is data that can be compared with the user's choice response and may serve as a basis for analyzing the user's actual interests and behavioral tendencies. Behavioral pattern information is first-party data that can be compared with the user's choice response and may be utilized to analyze the user's actual interests and behavioral tendencies. Behavioral pattern information may include data such as the user's actual consumption habits, service usage history, content consumption records, and device usage patterns. For example, a list of products purchased by the user at an online shopping mall, usage history of a recurring subscription service, search and click records in a mobile app, and a list of content viewed on a streaming platform may correspond to behavioral pattern information.

[0108] In addition, data collected from a user's smart devices can also serve as behavioral pattern information. For example, data such as heart rate changes, step counts, exercise records, and sleep patterns measured by wearable devices can be utilized as behavioral pattern information related to the user's health and lifestyle. Location information and app usage records from mobile devices may also be included, and patterns such as frequently visiting specific stores or using specific apps at specific times can also serve as behavioral pattern information.

[0109] This behavioral pattern information is not merely a single data point, but rather accumulated data over a certain period that can reflect users' long-term habits and tendencies. For example, if the types of products users consume vary depending on the season, or if their behavioral patterns differ between weekdays and weekends, these changes can also be included in the behavioral pattern information.

[0110] In a specific embodiment, the computing device (100) can determine multiple queries related to the user's interests by analyzing the user's previous responses and behavioral pattern information. The user's interests can be analyzed more precisely by including not only the responses explicitly provided by the user, but also behavioral pattern information that reflects the user's actual behavioral tendencies. The computing device (100) can dynamically configure appropriate queries to more accurately estimate the user's interests and tendencies.

[0111] For example, if a user responds, "I like coffee," but in actual consumption records (first-party data), they mainly purchase tea products, the computing device (100) may determine that the user's beverage preference may differ in certain situations. In this case, by taking into account the difference between the user's consumption pattern and the response, a customized query such as "Which do you prefer more in which situation, tea or coffee?" or "What beverage have you drunk most often recently?" can be constructed.

[0112] As another example, if a user responds, "I exercise more than 3 times a week," but the exercise frequency appears low in the exercise records of the wearable device, the computing device (100) may need to verify whether the user's exercise habits actually match the response. In this case, the user's exercise habits can be more clearly identified by adding questions such as, "What is the biggest motivating factor for exercising?" or "What is the biggest reason you cannot exercise often?"

[0113] In this way, the computing device (100) analyzes the user's previous responses and behavioral pattern information to determine multiple queries that consider actual interests and behavioral tendencies, rather than simply reflecting the user's explicit responses. Through this, the reliability of the obtained zero-party data can be improved. That is, zero-party data that more precisely reflects the user's actual preferences can be obtained.

[0114] Additionally, the step of creating a user interface and providing it to a user terminal (200) may include the step (S120) of selecting at least two response items from a group of response candidates based on the user's real-time context information, previous response, and behavior pattern information.

[0115] Real-time context information is data that reflects the user's current environment or situation and may include data such as the device type, network status, location information, and response speed at the time the user responds to a query. Through this, more appropriate response options can be provided depending on the situation the user is in.

[0116] The computing device (100) can comprehensively consider the user's real-time context information, previous responses, and behavioral pattern information to select at least two response items that the user is likely to select from a group of response candidates. This allows the user to respond more intuitively and improves the accuracy of zero-party data.

[0117] For example, when a user responds to a query about "preferred activities on weekends," if the user's previous response data indicates a preference for "outdoor activities," the computing device (100) may prioritize providing response items related to outdoor activities, such as "hiking" or "camping." On the other hand, if the user's behavioral pattern information shows a high number of indoor exercise records or high usage time of streaming services, response items such as "home training" or "watching movies" may be selected.

[0118] As another example, when a user responds to a query about the "most important factor when planning a trip," if the user's previous response confirms they "like winter" and their behavioral pattern data shows a high history of searching for winter travel destinations, response options such as "winter travel destination recommendations" or "ski resort options" can be provided. On the other hand, if the user's location data indicates multiple recent visits to beach areas, options such as "accommodation recommendations near the beach" or "surfing experience programs" can be selected.

[0119] In this way, the computing device (100) can select response items that are more likely to be selected by reflecting the user's personal characteristics and real-time context, rather than simply providing fixed response candidates. Through this, response data that is more aligned with the user's actual interests can be obtained, and ultimately, it can contribute to generating more precise personalized recommendation information.

[0120] In various embodiments, the computing device (100) can determine the number of response items based on the user's previous response data, behavioral pattern information, and real-time context information. The computing device (100) can dynamically adjust the number of response candidate groups by comprehensively analyzing the user's previous response data, behavioral pattern information, and real-time context information. Through this, user response fatigue can be reduced, while collecting more precise Zero Party Data to generate customized recommendation information.

[0121] Specifically, the computing device (100) can adjust the number of response candidates by analyzing the user's response history. To this end, the user's past response data is stored and managed, and the data is analyzed to determine whether a specific type of response selection is made repeatedly. For example, if the user consistently selects "Americano" in response to a "coffee" related query, the computing device (100) determines that the user's coffee preference is clear and can reduce the number of response candidates by excluding options that the user is unlikely to respond to. For instance, the method of providing options such as "Americano," "Latte," "Cappuccino," "Mocha," and "Cold Brew" can be reduced to major options such as "Americano" and "Latte" by reflecting the user's response pattern. On the other hand, if the user has a history of selecting various types of coffee, more diverse options can be maintained or new options can be added to identify detailed preferences.

[0122] Additionally, the computing device (100) can adjust the number of response candidates based on the user's behavioral pattern information and interests. To this end, consumption records, service usage history, content consumption patterns, etc. extracted from the user's First Party Data can be analyzed to determine whether there is a high preference for a specific brand or category. For example, if a record of continuously selecting products of a specific sports brand (Brand A) is confirmed in the user's purchase history, the computing device (100) determines that the user prefers Brand A and can exclude options that are not directly related to Brand A from the response candidates. For example, in a method of providing a selection of one of "Brand A," "Brand B," "Brand C," or "Brand D" in response to the query "Select a sneaker brand," if the user's behavioral pattern is closely related to Brand A, the response candidates can be reduced to two options: "Brand A" and "Brand B." On the other hand, if the user's behavioral pattern is not concentrated on a specific brand but selects various brands, the existing four response candidate groups can be maintained or new brand options can be added to analyze preferences more precisely.

[0123] Additionally, the computing device (100) can optimize the number of response candidates based on the user's real-time context information. To this end, the user's current device environment, network status, response speed, etc., can be detected in real time, and the optimal number of response candidates can be determined by reflecting this.

[0124] For example, when a user responds via a mobile device, input needs to be simple, so the number of response options can be reduced to optimize the response so that the user can respond more quickly. For instance, in response to the question "What activities do you prefer on weekends?", when a user responds using a smartphone, only three intuitively selectable response options, such as "walking," "watching movies," and "shopping," can be provided to enable a quick selection.

[0125] On the other hand, when users respond in a desktop environment, more granular choices are possible, allowing the pool of response candidates to be expanded to 5 to 7. For example, for the same question, options such as "reading," "exercising," "cooking," and "listening to music" can be added in addition to "walking," "watching movies," and "shopping" to identify more detailed preferences.

[0126] In addition, the number of response candidates can be optimized based on network conditions. For example, if a user responds in an unstable network environment (e.g., when the LTE connection is unstable while moving), the number of response candidates can be reduced to minimize data transmission. Conversely, if the user responds in a high-speed Wi-Fi environment, a more diverse range of response candidates can be provided. For instance, regarding the question "What is the most important factor when traveling?", if the network conditions are unstable, only two or three key response candidates, such as "cost" or "convenience," may be provided. On the other hand, in a stable network environment, more detailed response candidates such as "cost," "convenience," "activities," "local food," and "accommodation environment" can be provided.

[0127] As another example, response speed information can be used to adjust the pool of response options. If a user tends to make quick selections based on past responses, the pool of options can be minimized to encourage an intuitive choice. Conversely, if a user tends to respond slowly, a more diverse range of options can be provided to identify their detailed preferences. For instance, in response to the question "What food do you prefer for breakfast?", if the user exhibits a pattern of quick responses, simple options such as "bread," "rice," and "fruit" can be provided. On the other hand, if the user responds thoughtfully, a wider variety of choices can be offered, including "bread," "rice," "fruit," "salad," "yogurt," and "cereal."

[0128] In this way, the computing device (100) can collect zero-party data more effectively by comprehensively analyzing the user's past response patterns, behavioral pattern information, and real-time context information to dynamically adjust the number of response candidate groups. Through this, it is possible to provide the effect of generating more accurate personalized recommendation information while minimizing the user's response fatigue.

[0129] Additionally, the step of creating a user interface and providing it to a user terminal may include the step (S130) of creating a selection input screen including a plurality of determined queries and response items corresponding to each query.

[0130] The computing device (100) can dynamically configure a selection input screen that allows the user to respond intuitively based on previously determined question and response items. Considering the user's preferences and response patterns, the selection input screen can be provided in various forms, such as text-based, image-based, or icon-based, and can be optimized into an appropriate interface form according to the user's device environment.

[0131] For example, when a user responds using a smartphone, a selection input screen in the form of large buttons can be created to facilitate touch operation. For instance, in response to the question "What is your preferred activity on the weekend?", options such as "walking," "watching movies," and "shopping" can be provided along with icons to enable intuitive selection. On the other hand, when responding in a desktop environment, a list or card-based UI can be applied to allow a wider range of response options to be viewed at a glance.

[0132] As another example, image or video-based selection input screens can be provided to analyze user preferences more clearly. For instance, when a user selects a travel destination preference, queries such as "Beach vs. Mountain" or "City Travel vs. Nature Travel" can be presented with actual photos of the destinations rather than text, allowing the user to respond intuitively.

[0133] Additionally, the placement of response items can be optimized by reflecting information on users' response and behavioral patterns. For example, if a user has shown a tendency to prefer a specific option in past responses, that option can be placed at the top to speed up the response. For instance, if a user has previously selected "Americano" multiple times in response to the question "Please select your preferred beverage," the "Americano" option can be placed in the most prominent position to enable the user to respond intuitively.

[0134] According to one embodiment of the present invention, a computing device (100) can dynamically provide an optimal user interface by analyzing the environment and user response situation of a user terminal (200) in real time. Through this, the user's response experience can be optimized and support can be provided to enable more accurate Zero Party Data collection.

[0135] In a specific embodiment, the step of creating and providing a user interface to the user terminal may include receiving real-time environment information from the user terminal, determining which screen to provide among a selection input screen and a user input screen based on the real-time environment information, and providing the user interface including the determined screen to the user terminal.

[0136] In an embodiment, real-time environment information is information regarding the situation in which a response input occurs and the state of the user terminal, and may include at least one of device type information, network status information, location information, and response speed information.

[0137] For example, when a user responds while on the move using a mobile device, the computing device (100) can provide a fast and concise selection input screen by analyzing real-time environment information. On the other hand, when a user can respond with sufficient time in a desktop environment, a user input screen that allows for more detailed response input can be provided.

[0138] Additionally, network status information can be utilized to provide an optimal interface. For example, if a user responds in an environment where the network connection is unstable, the computing device (100) can provide a simple text-based selection input screen with a low data loading burden. On the other hand, in a high-speed network environment, an image or video-based interface can be utilized to provide a more immersive response experience.

[0139] Additionally, response speed information can be utilized to determine an appropriate user interface. For example, if a user has previously demonstrated fast response times, a concise selection input screen containing intuitively selectable response options can be provided. Conversely, if a user has previously shown cautious response patterns, a user input screen allowing for more free responses can be provided to collect detailed feedback.

[0140] According to one embodiment of the present invention, a computing device (100) may be configured to obtain more refined data by evaluating the reliability of collected zero-party data and analyzing consistency with previous responses of the same user to correct responses with low reliability. Through this, the user's response data is maintained more consistently, and customized recommendation information can be generated based on highly reliable data.

[0141] In a specific embodiment, referring to FIG. 5, a method for providing customized recommendations through the collection and analysis of Zero Party Data may include a step (S210) of analyzing the response pattern of Zero Party Data to calculate consistency information with the previous response of the same user.

[0142] In the embodiments, consistency information may be information for evaluating the logical continuity between a user's previous response data and the current response. Specifically, consistency information may be used to analyze how well the current response matches an existing pattern based on the same user's past response patterns.

[0143] For example, if a user responds that they "usually enjoy drinking coffee" and then responds that they "do not drink coffee" in a subsequent beverage preference question, the computing device (100) can detect the discrepancy between the two responses and reflect this in the consistency information. On the other hand, if the user consistently responds that they "like exercising" and then responds that they "mainly exercise on weekends" in subsequent questions, a high consistency score can be assigned.

[0144] Furthermore, consistency information can be utilized to analyze the contextual connectivity of responses. For example, if a user responds that they "like winter sports," a high consistency score can be assigned if they select "skiing" or "snowboarding" in the subsequent question regarding preferred activities. Conversely, if they make a choice unrelated to winter sports, such as "surfing," a relatively low consistency score may be assigned.

[0145] The computing device (100) can analyze the user's response pattern to evaluate consistency with the same user's previous responses. To do this, the computing device (100) can refer to the user's past response data to examine whether a specific response is logically consistent with existing responses. For example, if the user responded that they "prefer winter" in a previous question but responded that they "prefer summer" in a subsequent season-related question, the computing device (100) can detect this discrepancy.

[0146] In addition, it is possible to analyze whether the user's response maintains a consistent pattern. For example, if a user consistently responds that they prefer "healthy food" but responds that they "enjoy fast food" in response to a specific question, this may differ from the user's existing response pattern. In this way, the computing device (100) can produce consistency information based on the user's response pattern and can quantitatively evaluate the reliability of the response by assigning a consistency score.

[0147] Additionally, the method for providing customized recommendations through the collection and analysis of Zero Party Data may include a step (S220) of calculating reliability information of the response by comparing First Party Data and the Zero Party Data.

[0148] In the embodiments, reliability information may be information that evaluates how well a user's response matches actual behavioral data. Reliability information may be calculated by analyzing the correlation between the user's Zero Party Data (e.g., directly provided response) and First Party Data (e.g., actual behavioral data). Through this, it is possible to determine how well the user's subjective response matches objective data, and to perform correction or additional verification if necessary.

[0149] Specifically, the computing device (100) can evaluate the reliability of a response by comparing the user's first-party data and zero-party data. For example, if a user responds, "I mainly consume health foods," but there are many records of purchasing fast food or carbonated drinks in the actual purchase history (First Party Data) rather than health foods, the computing device (100) may evaluate the reliability of the response as low. On the other hand, if a user responds, "I regularly use the gym," and actual exercise-related app usage records or activity data from a smart device support this, the reliability of the response may be evaluated as high.

[0150] According to the embodiment, reliability information can be utilized to analyze the consistency between a user's response patterns and actual behavior. For example, if a user responded that they "travel abroad frequently" in the past, but there is little record of overseas travel in actual location data or flight reservation history, the reliability of that response may be lowered. On the other hand, if records of visiting multiple countries are confirmed in the user's location information, a high reliability score may be assigned.

[0151] Additionally, the method for providing customized recommendations through the collection and analysis of Zero Party Data may include the step (S230) of providing an additional user interface including a re-inquiry to the user terminal when the confidence score is below a preset threshold, and performing correction on the response based on the response to the additional user interface.

[0152] Specifically, the computing device (100) may perform an additional verification process to increase the accuracy of the response if the reliability score of the user's response is below a preset threshold. To this end, the computing device (100) may provide an additional user interface including a re-inquiry to the user terminal (200) and obtain a complementary response from the user to re-evaluate the reliability of the existing response.

[0153] For example, if a user responds, "I mainly consume organic food," but the frequency of purchasing organic products in First Party Data (actual purchase history) is low or non-existent, the computing device (100) may evaluate the reliability of the response as low. In this case, additional questions such as "Have you purchased organic food in the last 3 months?" or "What is the reason you have not purchased organic food?" may be provided through an additional user interface. If the user provides an additional response to this, the computing device (100) may evaluate consistency with the existing response and, if necessary, generate corrected data.

[0154] Additionally, the re-questioning can be customized by analyzing the user's response patterns. For example, if a user responds that "I exercise every day," but wearable device data or fitness app usage records do not support this, the computing device (100) can provide supplementary questions such as "What type of exercise do you do most often?" or "If there are days when you cannot exercise, what is the main reason?" Through this, it is possible to derive deeper insights into the user's behavior patterns beyond simple verification of response reliability.

[0155] Additionally, re-questions can be dynamically modified based on the user's response. For example, if a user responds that they "tend to eat at home rather than dining out," but their delivery app usage history reveals frequent dining out, the user's lifestyle can be analyzed more accurately through additional questions such as, "What type of food do you cook most often when eating at home?" or "What is your main reason for using delivery food?"

[0156] In this way, the computing device (100) evaluates reliability information to analyze how well the user's response data corresponds to actual behavior, and can perform a correction process for responses with low reliability or provide additional verification questions. Through this, a more sophisticated user profile can be generated and the accuracy of personalized recommendations can be improved.

[0157] Additionally, the method for providing customized recommendations through the collection and analysis of Zero Party Data may include a step (S240) of generating Zero Party Data by applying weights to responses based on consistency information.

[0158] According to an embodiment, the computing device (100) evaluates the degree of consistency between a user's response and existing response patterns and behavioral data, and can adjust the weight of zero-party data by reflecting the evaluation result. Through this, high-reliability responses are given higher weights to improve the accuracy of the analysis, and low-reliability responses are given lower weights or adjusted through the collection of complementary data.

[0159] For example, when a user responds, "I mainly exercise in the evening," if the exercise records of the wearable device or the usage history of the fitness app are actually concentrated in the evening, the computing device (100) may determine that the consistency of the response is high and assign a high weight. On the other hand, if the user's exercise records are mainly concentrated in the morning or afternoon, the reliability of the response may be evaluated as low, and when analyzing the data, a low weight may be assigned or an additional verification step may be performed.

[0160] Additionally, if consistency is maintained compared to the user's previous responses, the weight of the response may be increased. For example, if the user has responded several times in the past that "I enjoy outdoor activities on weekends" and there is a record of visiting a park or mountain every weekend in actual location data and social media activity history, the computing device (100) may determine that the user's response is reliable and increase the weight of the response. On the other hand, if the user previously responded that "I mainly spend time indoors" but in this response says "I mainly do outdoor activities," it may determine that there is a lack of consistency and lower the weight of the response or perform an additional verification procedure.

[0161] Additionally, consistency information can be dynamically reflected by considering adaptability to changes in the user's lifestyle. For example, by analyzing the user's dietary data over the past few months, the weight of a response can be increased if a pattern of consistently consuming a specific type of food is maintained, while the weight of a newly entered response can be gradually adjusted if a sudden change is detected after a certain point. Through this, it becomes possible to capture long-term behavioral changes and enable more precise data analysis by utilizing weights adjusted in real time.

[0162] In this way, the computing device (100) can generate more reliable zero-party data by evaluating the consistency between the user's response pattern and actual behavior data and applying weights to the response. Through this, it is possible to provide the effect of improving the accuracy of the analysis results and increasing the precision of personalized recommendation information.

[0163] In various embodiments, the method for providing customized recommendations through the collection and analysis of Zero Party Data may include the step of determining an incentive based on completeness, reliability information, and participation frequency corresponding to the response, and providing the determined incentive to a user terminal.

[0164] The computing device (100) can determine the optimal incentive by evaluating the user's response quality and participation pattern and provide it to the user terminal (200). To this end, the computing device (100) can establish criteria for providing incentives by comprehensively analyzing the completeness and reliability information of the response and the frequency of participation, and can continuously induce the user's data provision activities.

[0165] In an embodiment, completeness may be a factor for evaluating whether a response provided by a user meets predefined criteria. For example, if a user omits a required response in a question requiring a multiple-choice response, or enters a very short or meaningless response in a free-form response, the computing device (100) may evaluate the completeness of the response as low. On the other hand, if the user answers all required questions and writes a response including specific details, the completeness may be evaluated as high, and a high incentive may be provided accordingly.

[0166] In addition, reliability information can be reflected in incentive decisions by evaluating the extent to which a user's response aligns with first-party data and existing zero-party data. For example, if a user responds, "I exercise every day," but analysis of actual behavioral data reveals that there is almost no exercise record on the user's wearable device, the reliability score may decrease, and the size of the incentive may be adjusted. On the other hand, if the user's response shows high consistency with existing behavioral patterns and past responses, the reliability score increases, and consequently, a higher incentive may be provided.

[0167] Additionally, the frequency of participation can serve as a criterion for evaluating whether the user consistently provides responses over a certain period. For example, if the user provides responses regularly every day or every week, the computing device (100) recognizes this as an important factor contributing to long-term data collection and may gradually increase the reward level to induce continued participation. Conversely, if the user responds irregularly or does not provide responses for a long period, the frequency of providing incentives may be reduced or additional messages encouraging participation may be provided.

[0168] In various embodiments, the computing device (100) may apply differential incentive policies by taking into account the user's response quality and patterns. For example, users with high completeness and high reliability scores may receive incentives such as increased point accumulation rates, bonus rewards, and additional benefits, and users who consistently provide high-quality responses may be provided with special rewards (e.g., discount coupons, promotion codes, etc.). On the other hand, if reliability is low or response quality does not meet certain standards, the provision of incentives may be limited.

[0169] In this way, the computing device (100) can determine an incentive by comprehensively considering the user's response completeness, reliability information, and frequency of participation, thereby securing more reliable zero-party data and providing the effect of inducing long-term user participation.

[0170] Additionally, in the embodiment, the method for providing customized recommendations through the collection and analysis of Zero Party Data may include the step of dynamically adjusting the timing of the provision of the user interface based on the time at which the response is collected and the participation pattern.

[0171] Specifically, the computing device (100) can analyze the timing of when a user's response is provided and the participation pattern in real time to adjust the user interface so that it can be provided at an optimal time. To this end, the computing device (100) can establish a dynamic interface provision strategy based on the user's response frequency, previous response time zone, participation cycle, usage pattern, etc.

[0172] In the embodiment, the response frequency may indicate how often a user provides a response during a specific period. For example, if a user shows a pattern of responding at a specific time of day (e.g., 8:00 AM to 9:00 AM), the computing device (100) can learn the user's preferred time and adjust to provide the user interface at the same time. On the other hand, if the user's response frequency is inconsistent or irregular, the method of providing the interface to the user can be optimized by setting appropriate time intervals.

[0173] Additionally, the timing of interface provision can be adjusted by considering previous response times. For example, if a user shows a pattern of responding more frequently on specific days or at specific times (e.g., after lunchtime, evening time), the computing device (100) can provide a user interface tailored to those times. On the other hand, if a user tends to avoid responding at specific times, the timing of interface provision can be adjusted to avoid those times.

[0174] Additionally, the timing of providing the user interface can be optimized by considering the participation cycle. For example, if the user shows a pattern of responding once a day, the computing device (100) can be adjusted to provide the user interface once a day. On the other hand, if the user shows a pattern of responding once or twice a week, the cycle of providing the user interface can be increased or decreased according to that pattern.

[0175] In various embodiments, the interface provision strategy can be optimized by analyzing usage patterns. For example, if a user primarily responds via mobile devices during commuting hours, a simple selection input screen can be provided during that time, while a user input screen requiring more detailed input can be provided during the evening hours when there is more free time. Additionally, if a user frequently provides responses on specific days (e.g., weekends), the interface provision method can be adjusted by focusing on providing customized queries on weekends.

[0176] In this way, the computing device (100) learns the user's response provision pattern and provides a user interface at an optimal time, thereby maximizing the response rate and collecting more effective zero-party data.

[0177] According to one embodiment of the present invention, a method for providing customized recommendations through the collection and analysis of Zero Party Data may include the step (S200) of collecting Zero Party Data based on a response to a user interface.

[0178] In an embodiment, the step of collecting zero-party data may include the step of generating zero-party data by refining the response according to a predefined data structure, the step of encrypting the zero-party data using a pre-configured encryption algorithm, and the step of storing the encrypted zero-party data in a database and restricting data retrieval and utilization according to an access control policy.

[0179] To explain in more detail, the computing device (100) can generate zero-party data by converting collected response data into a structured data structure. To this end, the computing device (100) can map the response data to a predefined data field, remove unnecessary blanks or duplicate data, and perform tasks to consistently refine the response format. For example, if a user selects either "Summer" or "Winter," the computing device (100) can store this in a field called "Season_Preference" to maintain a consistent data structure. Additionally, if a user provides a subjective response, the computing device (100) can apply natural language processing (NLP) techniques to extract key keywords and classify the meaning of the response to convert it into structured data.

[0180] Refined zero-party data can be encrypted to enhance security. A computing device (100) can encrypt data using a pre-configured encryption algorithm (e.g., AES-256, RSA, SHA-256 hash algorithm, etc.) and ensure secure storage. For example, to prevent the user's response data from being stored in plaintext form, the computing device (100) can hash the user's response data or convert it into an encrypted form by applying a public key-based encryption technique. Through this, the stored data can be protected from unauthorized access and the user's personal information can be prevented from being leaked externally.

[0181] Encrypted zero-party data may be stored in a database according to a predefined access control policy. The computing device (100) may set access rights based on the user's level of consent and the purpose of data utilization, and may restrict access so that only specific users or systems can view or utilize the data. For example, if a user restricts the sharing of a specific type of data, the data may be configured to be accessible only to a specific analysis engine, and may be adjusted so that the data cannot be utilized by a system that provides personalized recommendations. As a specific example, if a user sets that they do not want health-related data to be used for marketing purposes when providing such data, the computing device (100) may reflect this and restrict access so that the data cannot be viewed by a personalized advertising system. On the other hand, access may be allowed to an analysis engine that provides health-related recommendation content, so that the data can be utilized for services such as health information preferred by the user or recommendations for customized nutritional supplements.

[0182] Additionally, when a user provides financial-related data, the computing device (100) may allow the financial service recommendation engine to analyze the data according to the user's settings, but may block access to the data from external third-party advertising platforms. For example, if the user sets the data to "allow use for consumption pattern analysis but not for advertising targeting," the consumption trend analysis system may view and process the data, but the customized financial product advertising system may not be able to utilize the data.

[0183] As another example, when a user uses a location-based service, if the user sets their location information to be shared only during specific time periods, the computing device (100) may restrict access to the user's location information to only during specific time periods according to a predefined access policy. For example, if the user sets the location data to be used between 9 AM and 6 PM, the location-based recommendation system can access the data only during that time period, and access is automatically blocked thereafter.

[0184] In this way, the computing device (100) can ensure transparency in data usage and maximize the protection of the user's privacy by applying an access control policy according to the user's level of consent and the purpose of data usage.

[0185] As described above, the computing device (100) can maintain data consistency, enhance security, and protect user privacy by refining zero-party data, encrypting and storing it, and applying access control.

[0186] According to one embodiment of the present invention, a method for providing customized recommendations through the collection and analysis of Zero Party Data may include the step (S300) of acquiring First Party Data corresponding to a user terminal.

[0187] In the embodiment, first-party data refers to data related to the actual behavior of a user, which may be collected directly by the computing device (100) or obtained from a linked system. First-party data may be collected based on the user's online and offline activities and may include information such as the user's purchase history, website visit history, application usage logs, search history, content consumption patterns, click data, and shopping cart history.

[0188] The computing device (100) can acquire first-party data in various ways. For example, when a user uses a website or application, the computing device (100) can automatically collect the user's page visit history, content consumption history, click events, etc. through a tracking script or log data analysis. Additionally, when a user logs in and uses a service, the computing device (100) can link the user's account information to analyze the user's activity history and store it as first-party data.

[0189] For example, on an e-commerce platform, a computing device (100) can obtain first-party data by tracking a user's purchase history, shopping cart addition history, payment status, etc. If a user has viewed a specific product multiple times but has not purchased it, the computing device (100) can perform customized marketing strategies, such as recommending related products or providing additional discount coupons based on the data.

[0190] Additionally, on a streaming platform, the computing device (100) can collect first-party data by analyzing information such as a list of content viewed by the user, viewing time, preference for specific genres, and frequency of content playback. For example, if the user repeatedly watches videos of a specific genre, the computing device (100) can recommend similar content based on the data or optimize a customized recommendation algorithm by reflecting the user's viewing patterns.

[0191] In the case of a mobile application, the computing device (100) can obtain first-party data by analyzing the user's app usage frequency, specific function usage patterns, push notification response, etc. For example, if the user frequently uses a specific function, the computing device (100) can provide a customized user experience by optimizing the user's interface or highlighting related functions.

[0192] In this way, the computing device (100) can acquire first-party data through direct interaction with the user or automatically collect it by analyzing the user's service usage data. The acquired first-party data can be analyzed together with zero-party data and utilized to create a more precise user profile and generate customized recommendation information.

[0193] According to one embodiment of the present invention, a method for providing customized recommendations through the collection and analysis of Zero Party Data may include the step (S400) of generating customized recommendation information through integrated analysis based on Zero Party Data and First Party Data.

[0194] In one embodiment, the integrated analysis based on zero-party data and first-party data can be performed through a machine learning-based integrated analysis model to interpret user data more precisely and provide customized recommendations that reflect the characteristics of individual users.

[0195] According to the embodiment, the integrated analysis model combines explicit user preference information obtained from zero-party data with user behavior patterns analyzed from first-party data to generate a more reliable user profile and derive optimal personalized recommendations based on said profile. By comparing and analyzing user response patterns with actual behavioral data, this model can be trained to reflect the user's interests and consumption tendencies in a practical way, rather than relying on simple response data.

[0196] In a specific embodiment, referring to FIG. 6, the step of generating customized recommendation information may include the step (S410) of building a training data set using zero-party data and first-party data obtained from a plurality of users.

[0197] The computing device (100) can refine zero-party data and first-party data collected by each user to form a training data set that can be used for model training. The training data set may include not only individual user data but also data collected from various user groups, and may be configured in a form that reflects the characteristics of a specific user group.

[0198] For example, if a user selects "summer" in response to the question "Which season do you prefer, summer or winter?", that response can be stored as zero-party data. Meanwhile, if "air conditioner purchase history," "summer sports activity search history," "summer travel booking history," etc. exist in the same user's first-party data, it may indicate that the user's actual behavior is focused on summer-related activities. The computing device (100) can combine these data to form a training data set.

[0199] Furthermore, during the process of constructing the training dataset, a data cleansing process can be performed by comparing users' zero-party and first-party data to select highly reliable data and remove outliers. For example, if a user responds that they "prefer coffee" but their actual purchase history shows they have bought more tea-related products than coffee, their response may be deemed inconsistent. In such cases, the data can be corrected or weights adjusted to be reflected in the training dataset.

[0200] Additionally, training datasets can be transformed into a form suitable for specific machine learning models. For example, when utilizing neural network models, collected data can be vectorized and converted so that it can be processed in the input layer. On the other hand, when utilizing decision tree models, training data can be organized by classifying it into specific categories.

[0201] As described above, the computing device (100) can build a highly reliable training data set by utilizing zero-party data and first-party data obtained from various users, and perform subsequent steps to generate customized recommendation information based thereon.

[0202] Additionally, the step of generating customized recommendation information may include the step (S420) of training a machine learning-based integrated analysis model for user segment allocation using a training data set.

[0203] The computing device (100) can analyze the characteristics of user groups based on a training data set and learn a machine learning-based integrated analysis model that reflects this. The model does not directly utilize individual user data, but rather identifies user groups (segments) that share specific patterns and performs the role of assigning new user data to appropriate segments.

[0204] For example, the computing device (100) can analyze the relationship between zero-party data and first-party data to automatically cluster users who share a specific pattern. Depending on the user's preferences and behavioral patterns, various segments such as a 'summer sports preference group', a 'winter travel preference group', and a 'health-oriented consumer group' can be derived.

[0205] A machine learning-based integrated analysis model can be trained using various algorithms. For example, a computing device (100) can apply a K-Means clustering algorithm to analyze user behavior data and automatically form user groups with similar characteristics. Through this, instead of analyzing individual user data individually, users who share specific behavioral patterns can be classified into a single segment, thereby enabling more efficient customized recommendations.

[0206] Furthermore, by utilizing Random Forest and Decision Tree-based models, it is possible to predict which segment users with specific characteristics are likely to belong to. For example, by comparing and analyzing user response patterns with actual behavioral data, it is possible to identify which group users exhibiting specific purchasing tendencies or interests are likely to belong to.

[0207] Additionally, by utilizing neural network-based classification models, various user data can be analyzed multidimensionally, enabling more precise segment assignment. For example, by comprehensively analyzing individual response data, behavioral data, and real-time contextual information, it is possible to go beyond simple pattern analysis and even predict users' potential interests.

[0208] In this way, the computing device (100) learns an integrated analysis model that learns the relationships between user data, thereby enabling it to classify a user into an appropriate segment when a new user is input. Through this, a foundation can be laid to provide more precise customized recommendation information in a subsequent stage.

[0209] Additionally, the step of generating customized recommendation information may include the step (S430) of processing zero-party data and first-party data as input to an integrated analysis model and assigning them to user segments.

[0210] A computing device (100) can utilize a pre-trained machine learning-based integrated analysis model to analyze a user's zero-party data and first-party data, and assign the data to an appropriate user segment. To this end, the computing device (100) provides the user's response patterns, behavioral data, and past interaction data as input to the model and evaluates whether the user belongs to the most suitable segment.

[0211] For example, if a user consistently selects responses related to a "healthy diet" and the first-party data actually contains many organic food purchase records, the computing device (100) can assign the user to the "health-oriented consumer" segment. On the other hand, if the user's response pattern is unclear or shows various interests, the probability of belonging to multiple segments can be analyzed, and the user can be preferentially assigned to the segment showing the highest relevance.

[0212] Additionally, real-time context information can be utilized as a correction factor after segment allocation and can play a role in more precisely adjusting segment classification to suit the user environment and situation.

[0213] For example, if the user's current location data is adjacent to a specific store, the computing device (100) can analyze whether the user frequently visits the store, whether there is a history of past purchases, etc., and assign additional weight to the "local brand preference" segment. If a pattern is found in which the user repeatedly visits a specific type of store within a specific area, an additional segment subdivided into "user with a strong tendency for area-based shopping" may be assigned.

[0214] In addition, segment allocation can be optimized by considering content consumption patterns based on device type and network conditions. For example, if a user primarily consumes short-form content (e.g., short videos, news summaries) when using mobile devices but tends to prefer long-form content (e.g., in-depth analysis articles, documentaries) in a desktop environment, the weights of the "mobile-centric consumer" or "desktop-centric consumer" segments can be dynamically adjusted according to the user's device environment.

[0215] Network conditions can also be correlated with users' content consumption and shopping patterns. For example, if a user consumes HD streaming content in a high-speed Wi-Fi environment but prefers low-resolution or audio-based content when using a mobile network while on the move, the "streaming heavy user" or "audio content preferr" segment can be dynamically adjusted based on that user's network connection status.

[0216] In this way, the computing device (100) can basically assign user segments based on zero-party data and first-party data, and then perform optimal segment classification by reflecting real-time context information. Through this, personalized recommendations considering the user's situational context become possible, and more sophisticated customized information can be provided.

[0217] Additionally, the step of generating customized recommendation information may include the step of generating user profile information (S440) based on assigned user segments and zero-party data.

[0218] The computing device (100) can combine the user segment allocation results and zero-party data to construct user profile information that more precisely reflects the individual tendencies of the user. User profile information is information for identifying the characteristics of individual users belonging to a specific segment in more detail, and can be generated by comprehensively reflecting the individual's explicit preferences and behavioral patterns.

[0219] For example, if a user is assigned to the "Premium Brand Preferred" segment, the computing device (100) may additionally reflect in the profile if the user indicates in their direct response (i.e., ZPD, zero-party data) that they prefer a specific brand. That is, even if they belong to the same "Premium Brand Preferred" segment, one user may prefer high-end fashion brands while another user may show a concentrated interest in premium electronics brands, so these differential characteristics may be reflected in the profile information.

[0220] Additionally, the computing device (100) can correct the reliability of the profile based on the consistency of the user's responses and the results of the analysis of behavioral patterns. For example, if a behavioral pattern is found in which the user frequently searches for or purchases a specific type of product, the profile may be updated to indicate a preference for that product family. On the other hand, if a discrepancy is detected between the responses and the actual behavioral patterns, the reliability of the information may be lowered, or the profile update may be withheld until complementary data is collected.

[0221] Additionally, the computing device (100) can dynamically update user profile information to reflect the user's interests that change over time. For example, if the fashion style changes due to seasonal factors or if the user intensively searches for specific brand products during a specific promotion period, the change in the user's preferences can be reflected in the profile.

[0222] As such, the user profile generation method of the present invention does not stop at simple segment allocation, but combines zero-party data and behavioral pattern data to enable an in-depth understanding of individual users, and based on this, enables the provision of more precise personalized recommendation information.

[0223] Additionally, the step of generating customized recommendation information may include the step of generating customized recommendation information based on user profile information (S450).

[0224] Specifically, the computing device (100) can be configured to generate recommendation information optimized for the user's interests and behavioral patterns by utilizing user profile information. For example, if the user profile includes attributes such as “preference for healthy food” or “preference for outdoor activities,” the user’s customized recommendation information can be optimized for recommendations regarding low-calorie foods, sports equipment, outdoor brands, etc.

[0225] Additionally, the computing device (100) can be configured to provide more sophisticated recommendation information by combining user profile information with various recommendation algorithms. For example, by combining a collaborative filtering method and a content-based filtering method, content preferred by user groups with similar tendencies can be recommended, or new items with characteristics similar to items previously preferred by the user can be recommended.

[0226] In addition, it can be configured to dynamically optimize recommendation information by reflecting real-time contextual information. For example, when a user is at a specific location (e.g., a shopping mall, a cafe, an airport, etc.), products or services likely to be consumed at that location can be recommended first. Similarly, by considering the user's device type (e.g., mobile, tablet, desktop) and network status, a simple recommendation list can be provided in a mobile environment, while recommendation information including more detailed product descriptions and reviews can be provided in a desktop environment.

[0227] Additionally, the computing device (100) may be configured to finely adjust recommendation information by taking into account the reliability and update history of user profile information. For example, if a user has recently shown interest in a specific product category, the recommendation weight of that category may be gradually increased while maintaining a balance with past preferences.

[0228] As described above, the present invention can optimize customized recommendation information based on user profile information by reflecting various algorithms and real-time context. Through this, a more sophisticated recommendation experience tailored to the individual user's interests and consumption patterns can be provided.

[0229] According to one embodiment of the present invention, a computing device (100) may be configured to perform integrated analysis based on zero-party data and first-party data to dynamically adjust segment allocation by reflecting the latest data of the user and to provide more precise customized recommendation information.

[0230] According to an embodiment, the step of generating customized recommendation information through integrated analysis based on zero-party data and first-party data may include the step of acquiring additional zero-party data corresponding to the user.

[0231] The computing device (100) can continuously collect new inputs along with existing zero-party data to reflect the user's latest interests and behavioral changes. To do this, it can provide additional queries through a user interface or actively request data based on the user's specific behavior.

[0232] For example, if a user previously responded that they preferred "summer," but the season changes over time, the computing device (100) can check the user's latest preference by providing an additional question, "What is your current preferred season?"

[0233] Additionally, additional zero-party data may be obtained based on specific events or user behaviors. For example, if a user clicks on a specific brand product multiple times or adds it to a shopping cart but does not purchase it, the computing device (100) may provide additional questions to check if interest in the brand is sustained.

[0234] The computing device (100) of the present invention acquires additional zero-party data to reflect changes in the user in real time, thereby providing more precise customized recommendation information.

[0235] Additionally, the step of generating customized recommendation information may include a step of generating change information by comparing additional zero-party data with user profile information.

[0236] The computing device (100) can compare the acquired additional zero-party data with existing user profile information to analyze whether there is a change in the user's interests and behavioral patterns. Through this, it can evaluate whether the user's preferences are maintained or whether new interests have emerged.

[0237] For example, if a user responded that they prefer "winter sports" in an initial response, but subsequently provided a consistently positive response to a query related to "summer sports," the computing device (100) may determine that the user's seasonal sports preference may have changed.

[0238] Additionally, change information can be weighted by considering the frequency or intensity of the response. For example, if a user consistently emphasizes their preference for a specific brand, change information can be reflected by increasing the importance of information related to that brand. On the other hand, for one-time changes in responses, the decision to reflect the change information can be made by considering reliability.

[0239] In this way, the computing device (100) compares additional zero-party data with an existing user profile to generate change information that evaluates whether and to what extent preferences have changed, and based on this, can more precisely optimize customized recommendation information.

[0240] In addition, the step of generating personalized recommendation information may include a step of updating user profile information based on the generated change information.

[0241] The computing device (100) can dynamically update user profile information based on generated change information. The change information can be used to adjust the existing profile by reflecting the user's new interests, changes in behavioral patterns, response reliability, etc.

[0242] For example, if a user initially responded that they prefer "classical music," but recently continuously selects content related to "jazz music," the computing device (100) can determine that the user's musical taste has changed and perform an update by lowering the weight of "classical music" and raising the weight of "jazz music" in the user profile.

[0243] Additionally, if a preference for a specific category is strengthened, the computing device (100) can add new attributes to an existing profile. For example, if a user repeatedly searches for or purchases products related to "healthy food," a "healthy food preference" attribute can be added to the user profile to improve the accuracy of related recommendation information.

[0244] Profile updates can be reflected gradually, taking into account the reliability of the change information. In other words, by performing updates based on consistent patterns over a certain period rather than short-term changes, it is possible to prevent the profile from being excessively altered by momentary choices.

[0245] The computing device (100) can continuously adjust the user profile according to the change information and thereby provide more precise personalized recommendation information.

[0246] Additionally, the step of generating customized recommendation information may include a step of evaluating the results of updating user profile information and determining whether to change the segment to which the user belongs.

[0247] After the user profile is updated, the computing device (100) evaluates whether the change matches the user's existing segment and, if necessary, can change the segment to which the user belongs.

[0248] For example, if a user initially belongs to the "general sports interest user" segment but a pattern of intensively searching for and purchasing sports equipment of a specific brand is identified in recent data, the computing device (100) can reassign the user to the "brand preference sports user" segment.

[0249] Additionally, if a user's behavioral change deviates from the criteria of a specific segment, the computing device (100) may release the user's existing segment assignment and find a more suitable segment or create a new segment. For example, if a user initially belonged to the "Vegetarian Interested User" segment but has consistently selected meat-based dishes in recent data, the user may be excluded from the existing segment or changed to a new segment such as "Balanced Diet User."

[0250] In the segment change process, not only updated user profile information but also comparative analysis with other user groups exhibiting similar behavioral patterns may be utilized. Rather than using a specific user's temporary behavioral change as the basis for the segment change, the computing device (100) may decide whether to change the segment by considering data accumulated over a certain period together with the patterns of other similar users.

[0251] In this way, the computing device (100) can provide more precise customized recommendation information by continuously evaluating the user profile and adjusting the segments to reflect the changing user behavior patterns and interests.

[0252] In summary, the computing device (100) of the present invention can be configured to provide more precise customized recommendation information by integrating and analyzing the user's zero-party data and first-party data. To this end, the computing device (100) can dynamically generate an interface by comprehensively considering the user's response pattern, behavior pattern information, and real-time context information, and can provide an optimal set of question and response candidates.

[0253] Furthermore, the reliability of collected data can be evaluated and response consistency analysis performed to correct data quality, enabling the construction of profiles that more accurately reflect user interests. Moreover, by updating user profiles in real-time based on the latest changes in user data and adjusting segment allocations, the precision of personalized recommendation information can be enhanced.

[0254] According to an embodiment of the present invention, this data analysis and segment adjustment process can be performed using a machine learning-based integrated analysis model, thereby comprehensively reflecting the user's explicit and implicit preferences. Furthermore, by providing an optimal interface and recommendation method considering the user's environmental factors (device type, network status, etc.), it is possible to minimize user response fatigue and provide the effect of enabling natural and effective data collection.

[0255] As a result, the computing device (100) of the present invention can implement a data collection and analysis system that optimizes the integrated analysis of zero-party data and first-party data, reflects changes in user behavior in real time, and provides personalized, customized recommendation information with higher accuracy.

[0257] The components of the present invention may be implemented as a program (or application) and stored on a medium to be executed in combination with a computer, which is hardware. The components of the present invention may be implemented as software programming or software elements, and similarly, embodiments may be implemented in programming or scripting languages ​​such as C, C++, Java, assembler, etc., including various algorithms implemented as combinations of data structures, processes, routines, or other programming configurations. Functional aspects may be implemented as algorithms executed on one or more processors.

[0258] Those skilled in the art will understand that the various exemplary logic blocks, modules, processors, means, circuits, and algorithm steps described in connection with the embodiments disclosed herein may be implemented by electronic hardware, various forms of programs or design code (referred to herein as “software”), or a combination of all such. To clearly illustrate this interoperability between hardware and software, various exemplary components, blocks, modules, circuits, and steps have been generally described above in relation to their functions. Whether such functions are implemented as hardware or software depends on the design constraints imposed on the specific application and the overall system. Those skilled in the art may implement the functions described in various ways for each specific application, but such implementation decisions should not be interpreted as being outside the scope of the invention.

[0259] The various embodiments presented herein may be implemented as methods, devices, or articles of manufacture using standard programming and / or engineering techniques. The term “article of manufacture” includes a computer program, carrier, or medium accessible from any computer-readable device. For example, computer-readable media include, but are not limited to, magnetic storage devices (e.g., hard disks, floppy disks, magnetic strips, etc.), optical discs (e.g., CDs, DVDs, etc.), smart cards, and flash memory devices (e.g., EEPROMs, cards, sticks, key drives, etc.). Additionally, the various storage media presented herein include one or more devices and / or other machine-readable media for storing information. The term “machine-readable media” includes, but is not limited to, wireless channels and various other media capable of storing, holding, and / or transmitting command(s) and / or data.

[0260] It should be understood that the specific order or hierarchy of steps in the presented processes is merely an example of exemplary approaches. It should be understood that, based on design priorities, the specific order or hierarchy of steps in the processes may be rearranged within the scope of the invention. The appended method claims provide various step elements in a sample order, but do not imply limitation to the specific order or hierarchy presented.

[0261] The description of the presented embodiments is provided so that any person skilled in the art may use or practice the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be applied to other embodiments without departing from the scope of the present invention. Thus, the present invention is not limited to the embodiments presented herein, but should be interpreted in the broadest possible scope consistent with the principles and novel features presented herein.

Claims

Claim 1 A method for providing customized recommendations through the collection and analysis of zero party data, comprising: a step of providing a user interface to a user terminal, the user interface including a plurality of queries and an answer input window for inputting an answer to each of the queries, wherein the user interface is performed on one or more processors of a computing device; a step of collecting zero party data (ZPD) based on a response to the user interface; a step of acquiring first party data (FPD) corresponding to the user terminal; and a step of generating customized recommendation information through integrated analysis based on the zero party data and the first party data. Claim 2 In claim 1, the user interface comprises: a selection input screen for obtaining a selectable response from a user of the user terminal, including the plurality of queries and the answer input window; and a user input screen for obtaining a direct response from the user; and the step of creating and providing the user interface to the user terminal comprises: determining a plurality of queries related to the user’s interests based on the user’s previous responses and behavioral pattern information; selecting at least two response items from a group of response candidates based on the user’s real-time context information, the previous responses, and the behavioral pattern information; and creating a selection input screen including the determined plurality of queries and response items corresponding to each query; wherein the behavioral pattern information is information that serves as a basis for analyzing actual interests and behavioral trends through comparison with the user’s selected responses, a method for providing customized recommendations through the collection and analysis of zero-party data. Claim 3 In paragraph 2, the step of creating and providing the user interface to the user terminal comprises: receiving real-time environment information from the user terminal; determining which screen to provide among the selection input screen and the user input screen based on the real-time environment information; and providing the user interface including the determined screen to the user terminal; wherein the real-time environment information is information regarding the situation in which a response input occurs and the state of the user terminal, and includes at least one of device type information, network status information, location information, and response speed information, a method for providing customized recommendations through the collection and analysis of zero-party data. Claim 4 A method for providing customized recommendations through the collection and analysis of zero-party data, comprising: a step of analyzing the response pattern of the zero-party data to calculate consistency information with the previous response of the same user; a step of comparing the first-party data and the zero-party data to calculate reliability information of the response; a step of providing an additional user interface including a re-inquiry to the user terminal when the reliability score is less than a preset threshold, and performing correction on the response based on the response to the additional user interface; and a step of generating the zero-party data by applying weights to the response according to the consistency information. Claim 5 A method for providing customized recommendations through the collection and analysis of zero-party data, wherein, in claim 1, the method comprises: a step of determining an incentive based on completeness, reliability information and participation frequency corresponding to the response, and providing the determined incentive to the user terminal; and a step of dynamically adjusting the timing of providing the user interface based on the time at which the response is collected and the participation pattern. Claim 6 A method for providing customized recommendations through the collection and analysis of zero-party data, wherein, in claim 1, the step of collecting zero-party data based on a response to the user interface comprises: a step of generating zero-party data by refining the response according to a predefined data structure; a step of encrypting the zero-party data using a pre-set encryption algorithm; and a step of storing the encrypted zero-party data in a database and restricting data retrieval and utilization according to an access control policy. Claim 7 In claim 1, the step of generating customized recommendation information through integrated analysis based on zero-party data and first-party data comprises: a step of constructing a training data set using zero-party data and first-party data obtained from multiple users; a step of training a machine learning-based integrated analysis model for user segment allocation using the training data set; a step of processing the zero-party data and first-party data as input to the integrated analysis model and allocating them to user segments; a step of generating user profile information based on the allocated user segments and zero-party data; and a step of generating customized recommendation information based on the user profile information; a method for providing customized recommendations through the collection and analysis of zero-party data. Claim 8 In claim 7, the step of generating customized recommendation information through integrated analysis based on the zero-party data and the first-party data comprises: a step of acquiring additional zero-party data corresponding to the user; a step of generating change information by comparing the additional zero-party data with the user profile information; a step of updating the user profile information based on the generated change information; and a step of evaluating the result of updating the user profile information to determine whether to change the segment to which the user belongs; a method for providing customized recommendations through the collection and analysis of zero-party data. Claim 9 A device comprising: a memory for storing one or more instructions; and a processor for executing the one or more instructions stored in the memory, wherein the processor performs the method of claim 1 by executing the one or more instructions. Claim 10 A computer program stored on a recording medium readable by a computer, combined with a computer which is hardware, to perform the method of claim 1.

Images