Method, device and system for providing customized distribution-related information and matching platform service using an artificial intelligence model based on analysis of user status information

An AI-based system personalizes real estate presale information and consultation, addressing inefficiencies by matching users with experts and enhancing consultation quality and user satisfaction through data analysis and real-time interaction.

KR103034316B1Active Publication Date: 2026-11-10ZERO ONE PARTNERS CO LTD
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
KR1020250068731
Authority / Receiving Office
KR · KR
Patent Type
Patents
Current Assignee / Owner
Filing Date
2025-05-27
Publication Date
2026-11-10
Estimated Expiration
2045-05-27

AI Technical Summary

Technical Problem

Existing real estate and presale information services lack user customization, providing inconsistent quality and separating information and consultation, leading to inefficient user experiences and lower sales success rates.

Method used

An AI-based system analyzes user data to provide personalized presale information, matches users with suitable experts, and facilitates real-time consultation through an AI model, utilizing K-Means Clustering and natural language processing to enhance consultation quality and efficiency.

Benefits of technology

The system optimizes housing complex recommendations based on user preferences, improves consultation quality, reduces unnecessary searches, and enhances user satisfaction by providing intuitive environmental insights.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 112025059295315-PAT00003_ABST
    Figure 112025059295315-PAT00003_ABST
Patent Text Reader

Abstract

One embodiment of the present invention relates to an artificial intelligence (AI)-based technology for providing user-customized information, and more specifically, to a method, apparatus, and system for providing user-customized presale-related information (customized presale information) and matching platform services utilizing an AI model based on user status information analysis, which analyzes various status information such as a user's residential conditions, preferred areas, and subscription information to automatically match an optimal presale complex and presale expert, thereby enabling real-time interaction between the user and the presale expert.
Need to check novelty before this filing date? Find Prior Art

Description

Technology Field

[0001] The following embodiments relate to AI-based user-customized information provision technology, and in particular, to a method, apparatus, and system for providing user-customized presale-related information (customized presale information) and matching platform services utilizing an AI model based on user status information analysis, which analyzes various status information such as a user's residential conditions, preferred areas, and subscription information to automatically match the optimal presale complex and presale expert, thereby enabling real-time interaction between the user and the presale expert. Background Technology

[0002] Existing real estate and presale information services generally tend to be limited to providing information in the form of one-way lists, and there was a problem in that they could not accurately reflect the specific circumstances of individual users (e.g., housing subscription account information, income level, housing preference conditions, etc.).

[0003] Recently, some platforms have started providing recommendation features based on user search terms or chatbot consultations, but the quality of individual consultations is still inconsistent and there is a lack of visual information, which limits the ability to convey the actual environment or prospects of the housing complex to users.

[0004] Furthermore, since consultation services and information provision were separated, users faced the inconvenience of having to request a separate consultation or switch to another platform even after verifying the information. This hindered the user experience and acted as a factor in lowering the sales success rate. Prior art literature

[0005] (Patent Document 0001) KR 10-2566017 B (Patent Document 0002) KR 10-2192536 B (Patent Document 0003) KR 10-2687276 B The problem to be solved

[0006] The problem that an embodiment of the present invention aims to solve is to provide a method, apparatus, and system for providing a user-customized presale information provision and matching platform service utilizing an artificial intelligence model based on user status information analysis, which, in order to overcome the limitations of conventional real estate / apartment presale information provision methods as described above, analyzes complex user data such as subscription-related information, financial status, family composition, and areas of interest using an AI model to provide presale information that matches the user's qualifications and preferences, connects the user with the most suitable presale expert, and activates a channel for real-time consultation. means of solving the problem

[0007] According to one embodiment, a device includes a processor, memory, a communication module, and a non-transient storage medium, and a method for providing user-customized information related to property sales and a matching platform service utilizing an artificial intelligence model based on user status information analysis, wherein the method is performed by the processor by executing a program stored in the non-transient storage medium, comprises: a step of receiving user information from a user terminal; a step of generating property sales information by inputting the user information into a first artificial intelligence model; and a step of transmitting the property sales information to the user terminal.

[0008] Additionally, the method further comprises: a step of matching sales expert information corresponding to the sales information to the user terminal; a step of transmitting the sales expert information to the user terminal; and a step of activating a real-time consultation channel between the sales expert terminal corresponding to the sales expert information and the user terminal; wherein the user information includes: a type of housing subscription account, a housing subscription account subscription period, a housing subscription account deposit amount, a period of non-homeownership, a family composition, the number of dependents, an income level, asset information, a region of interest, and residential preference conditions; and the sales information includes: information on the housing complex, priority, and a subscription schedule; and the step of activating the real-time consultation channel includes: a first consultation step of inputting the user information into a second artificial intelligence model to generate first consultation assistance information; a second consultation step of transmitting the user information and the first consultation assistance information to the sales expert terminal; a third consultation step of adding conversation content additionally input into the real-time consultation channel to the second artificial intelligence model to generate second consultation assistance information; and a fourth consultation step of transmitting the second consultation assistance information to the sales expert terminal. and a fifth consultation step that repeats the third consultation step and the fourth consultation step until consultation termination information is received from the user terminal or the sales expert terminal; wherein the third consultation step comprises: a step of inputting the additionally input conversation content into the third artificial intelligence model to transform it into a standard conversation model; a step of converting the standard conversation model into a first-1 vector; a step of generating a first-2 vector by multiplying the first-1 vector by a first weight; a step of reading existing conversation content corresponding to the user information; a step of converting the existing conversation content into a second-1 vector; a step of generating a second-2 vector by multiplying the second-1 vector by a second weight; and a step of calculating a third-1 vector by summing the first-2 vector and the second-2 vector.A step of reading a first matrix from a database, wherein the first conversation example model is the 'row' and 'type of housing subscription account, housing subscription account subscription period, housing subscription account payment amount, period of non-homeownership, family composition, number of dependents, income level, asset information, area of ​​interest, housing preference conditions, and standard words' are the 'columns'; a step of generating a third-2 vector by adding the values ​​of 'type of housing subscription account, housing subscription account subscription period, housing subscription account payment amount, period of non-homeownership, family composition, number of dependents, income level, asset information, area of ​​interest, and housing preference conditions' read from the user information to the third-1 vector; a step of generating a second matrix by adding the third-2 vector to the first matrix; a step of classifying the second matrix into first clusters based on a K-Means Clustering model; a step of extracting a second cluster, which is the first cluster containing the first-1 vector, among the first clusters; and a step of receiving an existing conversation proposal list from the sales expert terminal. The method may include the step of extracting a second conversation example model, which is one of the 'rows' included in the second cluster, excluding the 'rows' of the third-2 vector and the existing conversation suggestion list; and the step of designating the second conversation example model as the second consultation assistance information.

[0009] And, the user terminal is controlled according to a control method comprising: receiving a rectangle on a first map surrounding the region of interest; and designating the rectangle as the region of interest; and the step of generating the third-2 vector comprises: receiving the coordinates of each vertex of the rectangle constituting the region of interest, namely (x11, y11), (x11, y12), (x12, y11), and (x12, y12) (where x11 < x12 and y11 < y12); and substituting x3 = (x12 - x11) / 5 and y3 = (y12 - y11) / 5; The method may include the step of substituting x21 = x11, x22 = x11 + x3, x23 = x11 + 2*x3, x24 = x11 + 3*x3, x25 = x11 + 4*x3, x26 = x12, y21 = y11, y22 = y11 + y3, y23 = y11 + 2*y3, y24 = y11 + 3*y3, y25 = y11 + 4*y3, and y26 = y12; the step of calculating first coordinates, which are 36 grid point coordinates formed based on x21 to x26 and y21 to y26; and the step of substituting the 36 first coordinates into 36 pre-specified 'column' positions among the 'columns' of the 3-2 vector.

[0010] In addition, the step of generating the above-mentioned sales information comprises: a step of reading the complex address from the above-mentioned sales complex information; a step of reading the complex area information from the above-mentioned sales complex information; a step of reading a bird's-eye view 3D model from the above-mentioned sales complex information; a step of drawing a first circle passing outside the complex on a first map based on the complex address and complex area information; a step of drawing a second circle on the first map that is concentric with the first circle and has a diameter of 105% to 108% of the diameter of the first circle; a step of drawing a third circle on the first map that is concentric with the first circle and has a diameter of 115% to 120% of the diameter of the first circle; a step of drawing a first straight line, which is a vertical line passing through the center of the first circle, on the first map; and a step of drawing a second straight line, which is a horizontal line passing through the center of the first circle, on the first map. A step of drawing a third line and a fourth line on a first map, which are diagonals passing through the center of the first circle and forming a 45-degree angle with the vertical and horizontal lines; a step of extracting 1-1 intersection points, which are intersection points between the first circle and the first to fourth lines; a step of extracting 2-1 intersection points, which are intersection points between the second circle and the first to fourth lines; a step of extracting 3-1 intersection points, which are intersection points between the third circle and the first to fourth lines; a step of extracting 1-2 intersection points, which are points on the first to fourth lines that are closest to each of the 1-1 intersection points among points on the first map where a road view exists; a step of extracting 2-2 intersection points, which are points on the first to fourth lines that are closest to each of the 2-1 intersection points among points on the first map where a road view exists. A step of extracting 3-2 intersection points that are closest to each of the 3-1 intersection points and are points on the 1st to 4th straight lines among the points where a road view exists on the 1st map;A step of reading a 1-1 Road View screen from each of the 1-2 intersection points; a step of reading a 2-1 Road View screen from each of the 2-2 intersection points; a step of reading a 3-1 Road View screen from each of the 3-2 intersection points; a step of extracting a 2D bird's-eye view screen for each of the 8 directions of East, West, South, North, Northeast, Southeast, Southwest, and Northwest from the bird's-eye view 3D modeling; a step of calculating a first direction, which is the location of each of the 1-2 intersection points, based on the center of the 1st circle; a step of calculating a second direction, which is the location of each of the 2-2 intersection points, based on the center of the 1st circle; a step of calculating a third direction, which is the location of each of the 3-2 intersection points, based on the center of the 1st circle; a step of generating a 1-2 Road View screen by compositing the 2D bird's-eye view screen corresponding to the first direction onto each of the 1-1 Road View screens; The method may include: a step of generating a 2-2 Road View screen by synthesizing the 2D bird's-eye view screen corresponding to the second orientation onto each of the 2-1 Road View screens; a step of generating a 3-2 Road View screen by synthesizing the 2D bird's-eye view screen corresponding to the third orientation onto each of the 3-1 Road View screens; a step of designating the 1-1 Road View screens, 1-2 Road View screens, 2-1 Road View screens, 2-2 Road View screens, 3-1 Road View screens, and 3-2 Road View screens as expected bird's-eye view information; and a step of including the expected bird's-eye view information in the sales information.

[0011] Additionally, the user terminal is controlled according to a control method further comprising: a step of displaying a second map on the user terminal; a step of marking points capable of displaying a bird's-eye view based on the expected bird's-eye view information on the second map; a step of reading a fourth road view screen for a selected bird's-eye view request point among the points capable of displaying a bird's-eye view from the expected bird's-eye view information; and a step of displaying the fourth road view screen; wherein the step of reading the fourth road view screen comprises: a step of reading a 4-1 road view screen, which is an existing road view corresponding to the bird's-eye view request point; and a step of reading a 4-2 road view screen, which is a composite road view corresponding to the bird's-eye view request point; and the step of displaying the expected bird's-eye view information comprises: a step of extracting a first image which is the difference between the 4-2 road view screen and the 4-1 road view screen; a step of extracting a first shape which is the outline of the first image; a step of generating a second image by converting the first image into grayscale; and a step of calculating a first pixel number which is the maximum horizontal pixel number of the first image. A step of calculating a second pixel number, which is the maximum vertical pixel number of the first image; a step of calculating a third pixel number, which is the horizontal pixel number of the 4-2 road view screen; 0.2% to 0.0% of the third pixel number.The method may include: a step of calculating a steel frame spacing of 3%; a step of calculating the number of steel frames, which is a value obtained by dividing the number of first pixels by the steel frame spacing; a step of calculating an inter-floor spacing, which is a value obtained by dividing the number of second pixels by 8 to 12; a step of displaying the 4-1 road view screen; a step of displaying the 1st shape on the upper layer of the 4-1 road view screen; a step of applying a glint animation to the 1st shape; a step of displaying multiple black vertical lines spaced apart from each other multiple times on the inner side of the 1st shape, wherein all black vertical lines corresponding to the number of steel frames are displayed within 3 to 4 seconds; a step of generating 3rd images by cutting the 2nd image from the bottom at each 'pixel corresponding to the inter-floor spacing'; a step of displaying the 3rd images by stacking them one by one from the bottom on the upper layer of the 1st shape; and a step of displaying the 1st image on the upper layer of the 3rd image.

[0012] A device according to one embodiment may be combined with hardware and controlled by a computer program stored on a medium to execute the method of any one of the methods described above. Effects of the invention

[0013] According to one embodiment, an optimized housing complex is recommended based on the user's actual subscription conditions and preferences, thereby reducing unnecessary information search and increasing the subscription success rate.

[0014] In addition, through consultation content analysis and prediction models, the quality of expert responses to user inquiries can be improved, and consultation time and effort can be reduced.

[0015] Furthermore, by providing projected bird's-eye view information similar to the actual environment, it enables users to intuitively grasp the surrounding environment, thereby facilitating more reliable decision-making.

[0016] In addition, by utilizing user-specified regions of interest for consultation using an unsupervised learning classification model, it is possible to assist in providing expert responses suitable for the user's region of interest, and by utilizing gridded values ​​of the region of interest data, the accuracy of the region of interest during unsupervised learning classification can be improved.

[0017] In addition, by analyzing and clustering existing counseling data to provide counseling information based on similar cases, it is possible to increase the work efficiency of experts and improve user satisfaction.

[0018] In addition, by generating a virtual street view screen and displaying it on the user's device based on visual effects, it is possible to inspire anticipation for the presale among users. Brief explanation of the drawing

[0019] FIG. 1 is a schematic diagram showing a system for providing user-customized sales-related information and matching platform services using an artificial intelligence model based on user status information analysis according to an embodiment of the present invention. FIG. 2 is a flowchart illustrating a method for providing user-customized sales-related information and matching platform services using an artificial intelligence model based on user status information analysis according to an embodiment of the present invention. FIG. 3 is a diagram illustrating the third consultation step of a method for providing user-customized sales-related information and matching platform services using an artificial intelligence model based on user status information analysis according to an embodiment of the present invention. FIGS. 4 and 5 are drawings illustrating the step of generating sales information in a method for providing user-customized sales-related information and matching platform services using an artificial intelligence model based on user status information analysis according to an embodiment of the present invention. Specific details for implementing the invention

[0020] Hereinafter, embodiments are described in detail with reference to the attached drawings. However, various modifications may be made to the embodiments, and thus the scope of the patent application is not limited or restricted by these embodiments. It should be understood that all modifications, equivalents, and substitutions to the embodiments are included within the scope of the rights.

[0021] Specific structural or functional descriptions of the embodiments are disclosed for illustrative purposes only and may be modified and implemented in various forms. Accordingly, the embodiments are not limited to the specific disclosed forms, and the scope of this specification includes modifications, equivalents, or substitutions that fall within the technical concept.

[0022] Terms such as "first" or "second" may be used to describe various components, but these terms should be interpreted solely for the purpose of distinguishing one component from another. For example, the first component may be named the second component, and similarly, the second component may be named the first component.

[0023] When it is stated that a component is "connected" to another component, it should be understood that it may be directly connected to or coupled with that other component, or that there may be other components in between.

[0024] The terms used in the embodiments are for illustrative purposes only and should not be interpreted as intended to be limiting. Singular expressions include plural expressions unless the context clearly indicates otherwise. In this specification, terms such as "comprising" or "having" are intended to indicate the existence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0025] Expressions such as 'end part', 'both ends', 'one end', 'other end', and 'side end' of a component can be interpreted as referring to at least / any one of the end parts of that component.

[0026] In the description of the present invention, 'a method in which a device comprises a processor, a memory, a communication module, and a non-transient storage medium, and a program stored in the non-transient storage medium is executed by the processor,' the term 'method' may be interpreted as referring to the program stored in the non-transient storage medium itself or a part of the program.

[0027] The term 'Return' as used in the description of the present invention may refer to a result value being output, returned, or returned from a method, procedure, function, etc. used in a given program language / structure.

[0028] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as generally understood by those skilled in the art to which the embodiments pertain. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an ideal or overly formal sense unless explicitly defined in this application. For example, the term 'artificial intelligence model' may be selected from one or more of known general artificial intelligence models.

[0029] In addition, when describing with reference to the attached drawings, identical components are assigned the same reference numeral regardless of drawing symbols, and redundant descriptions thereof are omitted. In describing the embodiments, if it is determined that a detailed description of related prior art could unnecessarily obscure the essence of the embodiments, such detailed description is omitted.

[0030] The embodiments may be implemented in a form including a server computer and a database connected to the server.

[0031] According to one embodiment, a device includes a processor, memory, a communication module, and a non-transient storage medium, and a method for providing user-customized information related to property sales and a matching platform service utilizing an artificial intelligence model based on user status information analysis, wherein the method is performed by the processor by executing a program stored in the non-transient storage medium, comprises: a step of receiving user information from a user terminal; a step of generating property sales information by inputting the user information into a first artificial intelligence model; and a step of transmitting the property sales information to the user terminal.

[0032] The step of receiving user information from the above user terminal; involves receiving user information, which is information entered by the user, from a terminal (smart device) of a user who has subscribed to / connected to a platform according to an embodiment of the present invention in order to obtain distribution information.

[0033] User information may include general platform member information such as a user identification number (unique number), contact information, and social media ID, as well as 'user information related to the sale' described below.

[0034] In the step of generating sales information by inputting the above user information into the first artificial intelligence model, the user information is input into the first artificial intelligence model that has already been generated, and sales information that matches / corresponds to the user information can be curated and provided / recommended to the user.

[0035] The above-mentioned presale information may be a part of the information currently stored in the database. The above-mentioned first artificial intelligence model can extract / recommend presale information suitable for the above-mentioned user information from the database.

[0036] Since this AI technology for curation is a known technology, a detailed explanation of it itself is omitted, and the following description focuses on the parts that differ from conventional technology.

[0037] In the step of transmitting the above-mentioned distribution information to the user terminal, the generated / extracted distribution information is transmitted to the user terminal and provided to the user.

[0038] Additionally, the method further comprises: a step of matching sales expert information corresponding to the sales information to the user terminal; a step of transmitting the sales expert information to the user terminal; and a step of activating a real-time consultation channel between the sales expert terminal corresponding to the sales expert information and the user terminal; wherein the user information includes: a type of housing subscription account, a housing subscription account subscription period, a housing subscription account deposit amount, a period of non-homeownership, a family composition, the number of dependents, an income level, asset information, a region of interest, and residential preference conditions; and the sales information includes: information on the housing complex, priority, and a subscription schedule; and the step of activating the real-time consultation channel includes: a first consultation step of inputting the user information into a second artificial intelligence model to generate first consultation assistance information; a second consultation step of transmitting the user information and the first consultation assistance information to the sales expert terminal; a third consultation step of adding conversation content additionally input into the real-time consultation channel to the second artificial intelligence model to generate second consultation assistance information; and a fourth consultation step of transmitting the second consultation assistance information to the sales expert terminal. and a fifth consultation step that repeats the third consultation step and the fourth consultation step until consultation termination information is received from the user terminal or the sales expert terminal; wherein the third consultation step comprises: a step of inputting the additionally input conversation content into the third artificial intelligence model to transform it into a standard conversation model; a step of converting the standard conversation model into a first-1 vector; a step of generating a first-2 vector by multiplying the first-1 vector by a first weight; a step of reading existing conversation content corresponding to the user information; a step of converting the existing conversation content into a second-1 vector; a step of generating a second-2 vector by multiplying the second-1 vector by a second weight; and a step of calculating a third-1 vector by summing the first-2 vector and the second-2 vector.A step of reading a first matrix from a database, wherein the first conversation example model is the 'row' and 'type of housing subscription account, housing subscription account subscription period, housing subscription account payment amount, period of non-homeownership, family composition, number of dependents, income level, asset information, area of ​​interest, housing preference conditions, and standard words' are the 'columns'; a step of generating a third-2 vector by adding the values ​​of 'type of housing subscription account, housing subscription account subscription period, housing subscription account payment amount, period of non-homeownership, family composition, number of dependents, income level, asset information, area of ​​interest, and housing preference conditions' read from the user information to the third-1 vector; a step of generating a second matrix by adding the third-2 vector to the first matrix; a step of classifying the second matrix into first clusters based on a K-Means Clustering model; a step of extracting a second cluster, which is the first cluster containing the first-1 vector, among the first clusters; and a step of receiving an existing conversation proposal list from the sales expert terminal. The method may include the step of extracting a second conversation example model, which is one of the 'rows' included in the second cluster, excluding the 'rows' of the third-2 vector and the existing conversation suggestion list; and the step of designating the second conversation example model as the second consultation assistance information.

[0039] In the step of matching sales expert information corresponding to the sales information to the above user terminal, a sales expert already registered in a database to correspond to the sales information is matched with the user.

[0040] The term "sales expert information" can be interpreted as referring to a sales expert account (sales expert terminal).

[0041] For example, if the presale information matched to the user information includes Apartment A and Apartment B, the presale expert in charge of Apartment A can be matched with the presale expert in charge of Apartment B.

[0042] In the step of transmitting the sales expert information to the user terminal; and the step of activating a real-time consultation channel between the sales expert terminal corresponding to the sales expert information and the user terminal, the sales expert information matched to the user (name, history / experience, assigned sales properties, etc.) is transmitted to the user terminal, and a real-time consultation channel with the corresponding sales expert is activated.

[0043] Here, a real-time consultation channel can be any one of real-time chat, voice call, video call, voice conference, or video conference.

[0044] Activating a real-time consultation channel may refer to opening a chat room, connecting a call, etc.

[0045] The above user information may include at least one of: type of housing subscription account, housing subscription account subscription period, housing subscription account deposit amount, period of non-homeownership, family composition, number of dependents, income level, asset information, area of ​​interest, and housing preference conditions.

[0046] The above sales information may include: information on the sales complex, priority, and subscription schedule.

[0047] The sales complex information section may include information such as the area, number of households, location (address; coordinates), project scale, and bird's-eye view of the complex planned for each sales project.

[0048] The above priority may refer to a priority designated in correspondence with the above-mentioned information on the housing complex (such as whether it corresponds to the first priority under certain conditions).

[0049] The step of activating the above real-time consultation channel may include a function / process / task for suggesting appropriate / necessary response phrases or information, etc., to the consultation user on the sales expert terminal during the consultation.

[0050] In the first counseling step of generating first counseling assistance information by inputting the above user information into a second artificial intelligence model, the first counseling assistance information is generated based solely on the user information.

[0051] Here, the second artificial intelligence model mentioned above may consist of a series of steps including the K-Means Clustering model described below.

[0052] In the second consultation step of transmitting the user information and the first consultation assistance information to the sales expert terminal, the proposed first consultation assistance information is transmitted to the sales expert along with the user information, and the user information and the first consultation assistance information are displayed on the sales expert terminal.

[0053] The sales expert may select any of the contents in the first consultation support information to respond, or may write the response themselves.

[0054] In the third consultation step of generating second consultation assistance information by adding conversation content additionally entered into the real-time consultation channel to the second artificial intelligence model, the conversation content received from the user terminal is added to the second artificial intelligence model to newly generate second consultation assistance information.

[0055] In the fourth consultation step of transmitting second consultation assistance information to the above-mentioned sales expert terminal, the second consultation assistance information generated in real time by reflecting the user's conversation content is transmitted back to the sales expert terminal, and the second consultation assistance information is additionally displayed on the sales expert terminal.

[0056] In the fifth consultation step, which repeats the third consultation step and the fourth consultation step until consultation termination information is received from the user terminal or the sales expert terminal; the process described above is repeated until the real-time consultation between the user and the sales expert is terminated, so that whenever the user inputs any conversation, second consultation assistance information reflecting it is generated and displayed on the sales expert terminal.

[0057] In the step of inputting the additionally input conversation content into the third artificial intelligence model and transforming it into a standard conversation model, the conversation content entered by the user (additionally input conversation content) is transformed into a standard conversation model in the third artificial intelligence model, which is a natural language processing model.

[0058] Here, the standard conversation model refers to standardized words for key terms related to presales.

[0059] For example, terms such as 'Housing Subscription Comprehensive Savings Account', 'Housing Subscription Account', and 'Subscription Account' all have the same meaning, so they can be converted or modified into the standardized term 'Housing Subscription Comprehensive Savings'.

[0060] In the third AI model, particles or other unnecessary phrases / words are removed from the input conversation, and meaningful words are converted into standardized (pre-specified) words, and the converted data becomes the standard conversation model.

[0061] Natural language processing models of this type can follow known technologies.

[0062] In the step of converting the above standard conversation model into a first-1 vector, a first-1 vector is generated having the frequency of occurrence of key words included in the conversation content as values.

[0063] Vector 1-1 is a vector containing indices for all words used in the analysis of sales conversation / information data, and can have the frequency of occurrence of words extracted or transformed from keywords included in the conversation content entered by the user as the vector's 'value'.

[0064] For this process, the database stores a vector structure in which all standard words are used as 'column' indices.

[0065] For example, if standard words included in the user's conversation content, 'I have paid into the Housing Subscription Savings Account for 36 months and am interested in a presale of over 1,000 households in City A. My spouse also has a Housing Subscription Account; does that help?' are extracted, it can take the form of 'Housing Subscription Savings Account, Months of Payment, City A, Number of Households, Spouse', and if these words are the 'column' indices of the vector '35, 73, 155, 102, 40', then the 1-1 vector can be produced as [..., 1 (Housing Subscription Savings Account at index 35 appears twice), ..., 1 (Spouse at index 40 appears once), ..., 1 (Months of Payment at index 73 appears once), ..., 1 (Number of Households at index 102 appears once), ..., 1 (City A at index 155 appears once), ...].

[0066] In the step of generating a first-2 vector by multiplying the first-1 vector by a first weight; in order to generate second counseling assistance information that reflects the current conversation content more than the existing conversation content, the first-1 vector, which is a vector calculated from the most recent conversation, is multiplied by a first weight to increase the value of the first-1 vector (frequency of occurrence of words).

[0067] In the step of reading existing conversation content corresponding to the above user information; and the step of converting the existing conversation content into a 2-1 vector; the 'existing conversation content,' which is a standard conversation model accumulated and stored during consultation / conversation with the user up to that point, is converted into a 2-1 vector.

[0068] For example, if the user used words such as 'spouse, wife, spouse, etc.' a total of 7 times during this consultation, the value of the 2-1 vector is calculated as [..., 7 (spouse at index 40 appeared 7 times), ...].

[0069] In the step of generating a second-2 vector by multiplying the second-1 vector by a second weight, the second weight specified as a value less than 1 (e.g., 0.7) is multiplied by the second-1 vector to reduce the importance of the existing conversation content.

[0070] In the step of calculating the 3-1 vector by summing the 1-2 vector and the 2-2 vector, the 3-1 vector is calculated as a value that reflects both the most recently newly received conversation content from the user and the existing conversation content.

[0071] In the step of reading a first matrix from the database, where the first conversational example model is the 'row' and 'type of housing subscription account, housing subscription account subscription period, housing subscription account deposit amount, period of non-homeownership, family composition, number of dependents, income level, asset information, area of ​​interest, housing preference conditions, and standard words' are the 'columns', the first matrix, which is data already stored in the database, is read.

[0072] The first matrix is ​​exemplified in Fig. 3 (a).

[0073] The above-mentioned first conversation example model may refer to model answers that are already entered according to each case or product, etc. (unique model answer numbers such as QS10029 in Fig. 3 (a)).

[0074] The type of housing subscription account is a value that distinguishes between the comprehensive housing subscription savings account currently available for subscription as of 2025 and the old housing subscription savings, housing subscription deposits, and housing subscription installments.

[0075] For example, other types of the previous generation are designated with a single digit, and the current generation Housing Subscription Savings Account can be designated with a two-digit number.

[0076] The subscription period (months), subscription deposit amount (ten thousand won), period of non-homeownership (months), number of dependents (persons), and income level (ten thousand won) may each be numerical values ​​in the corresponding unit.

[0077] Information such as family composition (multiple children, newlyweds, etc.), areas of interest (Seoul, OO City in Gyeonggi-do, etc.), and residential preference conditions (large complex, school district, etc.) can each be designated to correspond to a specific code in the same way as the data on the types of housing subscription accounts described above.

[0078] For example, a user who is subscribed to the Housing Subscription Savings Account, is a multi-child household but not a newlywed couple, whose areas of interest are Seoul Metropolitan City and OO City in Gyeonggi-do, and who does not prefer large complexes but values ​​school districts, may have data (vector) in the form of Fig. 3 (b).

[0079] In the step of generating a 3-2 vector by adding values ​​read from the user information, such as 'type of housing subscription account, housing subscription account subscription period, housing subscription account deposit amount, period of non-homeownership, family composition, number of dependents, income level, asset information, area of ​​interest, and housing preference conditions,' to the 3-1 vector above, the values ​​read from the user information are combined with the 3-1 vector according to the 'column index identical to the 1 matrix.'

[0080] If the 3-1 vector is a 300-dimensional vector and the user information is a 35-dimensional vector, then the 3-2 vector becomes a 335-dimensional vector.

[0081] In other words, when the 3-1 vector is combined with user information of the form shown in Fig. 3 (b), the 3-2 vector of the form shown in any row of Fig. 3 (a) is generated.

[0082] In the step of generating a second matrix by adding a third-2 vector to the first matrix, the second matrix is ​​generated by adding the third-2 vector, which is data in which the user's conversation content and user information are combined, to the first matrix read from the database.

[0083] For example, if the first matrix was a matrix with 45531 'rows', the second matrix would be a matrix with 45532 'rows'.

[0084] In the step of classifying the above second matrix into first clusters based on a K-Means Clustering model, the second matrix is ​​classified using an unsupervised learning model, specifically a K-Means Clustering model.

[0085] The K-Means Clustering model itself follows the disclosed technology.

[0086] In the step of extracting a second cluster, which is a first cluster containing the first-1 vector among the first clusters above, the second cluster containing the first-1 vector, which is user data, is found among a plurality of first clusters classified from the second matrix.

[0087] In the step of receiving an existing conversation suggestion list from the above-mentioned sales expert terminal, an existing conversation suggestion list (a list of conversation example code numbers), which is data recording conversations that the sales expert has already mentioned (conversed) with the user, is read.

[0088] In the step of extracting a second conversation example model, which is one of the 'rows' included in the second cluster excluding the 'third-2 vector and the existing conversation proposal list', conversation examples included in the second cluster that have never been mentioned by a sales expert are extracted.

[0089] For example, in FIG. 3 (a), QS10029 to QS10036 and the 3-2 vector are extracted as the 2nd cluster, and if QS10033 and QS10036 are included in the existing conversation suggestion list, the 2nd conversation example model can be extracted as QS10029, QS10030, QS10031, QS10032, QS10034, and QS10035.

[0090] In the step of designating the above-mentioned second conversation example model as the above-mentioned second counseling assistance information; the extracted data is designated as the second counseling assistance information.

[0091] The designated second consultation assistance information is transmitted to and displayed on the sales expert's terminal, and the sales expert can select a desired one from the suggested conversation examples (second conversation example models), or if none of the suggested conversation examples are to their liking, they can continue the consultation by typing the conversation themselves.

[0092] And, the user terminal is controlled according to a control method comprising: receiving a rectangle on a first map surrounding the region of interest; and designating the rectangle as the region of interest; and the step of generating the third-2 vector comprises: receiving the coordinates of each vertex of the rectangle constituting the region of interest, namely (x11, y11), (x11, y12), (x12, y11), and (x12, y12) (where x11 < x12 and y11 < y12); and substituting x3 = (x12 - x11) / 5 and y3 = (y12 - y11) / 5; The method may include the step of substituting x21 = x11, x22 = x11 + x3, x23 = x11 + 2*x3, x24 = x11 + 3*x3, x25 = x11 + 4*x3, x26 = x12, y21 = y11, y22 = y11 + y3, y23 = y11 + 2*y3, y24 = y11 + 3*y3, y25 = y11 + 4*y3, and y26 = y12; the step of calculating first coordinates, which are 36 grid point coordinates formed based on x21 to x26 and y21 to y26; and the step of substituting the 36 first coordinates into 36 pre-specified 'column' positions among the 'columns' of the 3-2 vector.

[0093] The above user terminal may be controlled according to a control method comprising: receiving a rectangle on a first map surrounding the area of ​​interest; and designating the rectangle as the area of ​​interest.

[0094] In other words, the user can input / select an area of ​​interest in a rectangular shape by dragging one side of the first map displayed on the user terminal.

[0095] The step of generating the above 3-2 vector involves generating a vector portion of the form shown in FIG. 3 (c). The remaining portion of the above-described 3-2 vector is indicated as '...' in FIG. 3 (c).

[0096] In the step of receiving the coordinates of each vertex of the rectangle constituting the region of interest, namely (x11, y11), (x11, y12), (x12, y11), and (x12, y12) (where x11 < x12 and y11 < y12); the coordinates of the four vertices constituting the rectangle are each received.

[0097] In expressions such as 'x1' used in the description of the present invention, the number placed to the right of the alphabet may be interpreted as a subscript of the corresponding alphabet. For example, x1 is x_1( It can be interpreted as ).

[0098] In the step of substituting x3 = (x12 - x11) / 5 and y3 = (y12 - y11) / 5, x3 and y3, which are values ​​for dividing the rectangular area into 5 equal parts, are calculated (the calculated values ​​are assigned to the variables x3 and y3).

[0099] In the step of substituting x21 = x11, x22 = x11 + x3, x23 = x11 + 2*x3, x24 = x11 + 3*x3, x25 = x11 + 4*x3, x26 = x12, y21 = y11, y22 = y11 + y3, y23 = y11 + 2*y3, y24 = y11 + 3*y3, y25 = y11 + 4*y3 and y26 = y12, the coordinates of the grid points are calculated.

[0100] In the step of calculating first coordinates, which are 36 grid point coordinates formed based on the above x21 to x26 and y21 to y26, 36 points having the calculated 6 x-coordinates and 6 y-coordinates as grid point coordinates are designated as the first coordinates.

[0101] In the step of substituting the 36 first coordinates into the 36 pre-specified 'column' positions among the 'columns' of the 3-2 vector, the 36 grid point coordinates of the region of interest specified by the user are included in the values ​​of the 3-2 vector calculated according to the user information.

[0102] In addition, the step of generating the above-mentioned sales information comprises: a step of reading the complex address from the above-mentioned sales complex information; a step of reading the complex area information from the above-mentioned sales complex information; a step of reading a bird's-eye view 3D model from the above-mentioned sales complex information; a step of drawing a first circle passing outside the complex on a first map based on the complex address and complex area information; a step of drawing a second circle on the first map that is concentric with the first circle and has a diameter of 105% to 108% of the diameter of the first circle; a step of drawing a third circle on the first map that is concentric with the first circle and has a diameter of 115% to 120% of the diameter of the first circle; a step of drawing a first straight line, which is a vertical line passing through the center of the first circle, on the first map; and a step of drawing a second straight line, which is a horizontal line passing through the center of the first circle, on the first map. A step of drawing a third line and a fourth line on a first map, which are diagonals passing through the center of the first circle and forming a 45-degree angle with the vertical and horizontal lines; a step of extracting 1-1 intersection points, which are intersection points between the first circle and the first to fourth lines; a step of extracting 2-1 intersection points, which are intersection points between the second circle and the first to fourth lines; a step of extracting 3-1 intersection points, which are intersection points between the third circle and the first to fourth lines; a step of extracting 1-2 intersection points, which are points on the first to fourth lines that are closest to each of the 1-1 intersection points among points on the first map where a road view exists; a step of extracting 2-2 intersection points, which are points on the first to fourth lines that are closest to each of the 2-1 intersection points among points on the first map where a road view exists. A step of extracting 3-2 intersection points that are closest to each of the 3-1 intersection points and are points on the 1st to 4th straight lines among the points where a road view exists on the 1st map;A step of reading a 1-1 Road View screen from each of the 1-2 intersection points; a step of reading a 2-1 Road View screen from each of the 2-2 intersection points; a step of reading a 3-1 Road View screen from each of the 3-2 intersection points; a step of extracting a 2D bird's-eye view screen for each of the 8 directions of East, West, South, North, Northeast, Southeast, Southwest, and Northwest from the bird's-eye view 3D modeling; a step of calculating a first direction, which is the location of each of the 1-2 intersection points, based on the center of the 1st circle; a step of calculating a second direction, which is the location of each of the 2-2 intersection points, based on the center of the 1st circle; a step of calculating a third direction, which is the location of each of the 3-2 intersection points, based on the center of the 1st circle; a step of generating a 1-2 Road View screen by compositing the 2D bird's-eye view screen corresponding to the first direction onto each of the 1-1 Road View screens; The method may include: a step of generating a 2-2 Road View screen by synthesizing the 2D bird's-eye view screen corresponding to the second orientation onto each of the 2-1 Road View screens; a step of generating a 3-2 Road View screen by synthesizing the 2D bird's-eye view screen corresponding to the third orientation onto each of the 3-1 Road View screens; a step of designating the 1-1 Road View screens, 1-2 Road View screens, 2-1 Road View screens, 2-2 Road View screens, 3-1 Road View screens, and 3-2 Road View screens as expected bird's-eye view information; and a step of including the expected bird's-eye view information in the sales information.

[0103] The step of generating the above-mentioned sales information involves: combining an aerial view with an actual street view to predict what the street view will look like when the apartment is completed, and providing this so that users can visually verify it.

[0104] In the step of reading the complex address from the above-mentioned complex information, the complex address is read to determine where the corresponding complex is located on the map at specific points / coordinates.

[0105] In the step of reading complex area information from the above-mentioned complex information, the complex area information included in the above-mentioned complex information is read.

[0106] In the step of reading a bird's-eye view 3D model from the above-mentioned housing complex information, the bird's-eye view 3D model file of the corresponding housing complex included in the above-mentioned housing complex information is read.

[0107] In the step of drawing a first circle passing outside the complex on a first map based on the complex address and complex area information, the radius of the complex to be sold is estimated based on the complex area information and the complex address is centered thereon, and a first circle passing outside the complex is drawn.

[0108] In the step of drawing a second circle on the first map that is concentric with the first circle and has a diameter of 105% to 108% of the diameter of the first circle, a second circle that is larger and is positioned outside the first circle is drawn.

[0109] In the step of drawing a third circle on the first map that is concentric with the first circle and has a diameter of 115% to 120% of the diameter of the first circle, a third circle is drawn that is larger and is positioned outside the second circle.

[0110] In the step of drawing a first straight line, which is a vertical line passing through the center of the first circle, on the first map; the step of drawing a second straight line, which is a horizontal line passing through the center of the first circle, on the first map; and the step of drawing a third straight line and a fourth straight line, which are diagonals passing through the center of the first circle and forming a 45-degree angle with the vertical line and the horizontal line, on the first map, four straight lines are drawn that pass through the first to third circles and intersect at the center of the first circle.

[0111] In the step of extracting the first-1 intersection points, which are the intersection points between the first circle and the first to fourth lines, the first-1 intersection points, which are the intersection points between the four lines and the first circle, are extracted.

[0112] In the step of extracting the 2-1 intersection points, which are the intersection points between the 2nd circle and the 1st to 4th lines, the 2-1 intersection points, which are the intersection points between the four lines and the 2nd circle, are extracted.

[0113] In the step of extracting the 3-1 intersection points, which are the intersection points between the 3rd circle and the 1st to 4th lines, the 3-1 intersection points, which are the intersection points between the four lines and the 3rd circle, are extracted.

[0114] Among the points where a road view exists on the first map, the step of extracting first-2 intersection points that are closest to each of the first-1 intersection points and are points on the first to fourth straight lines; among the points where a road view exists on the first map, the step of extracting second-2 intersection points that are closest to each of the second-1 intersection points and are points on the first to fourth straight lines; and among the points where a road view exists on the first map, the step of extracting third-2 intersection points that are closest to each of the third-1 intersection points and are points on the first to fourth straight lines; wherein, among the points where a road view exists on the first map, the points where road views exist closest to each of the first-1 intersection points, second-1 intersection points, and third-1 intersection points are extracted.

[0115] In other words, 8 first-2 intersections, 8 second-2 intersections, and 8 third-2 intersections are extracted.

[0116] In the step of reading a 1-1 road view screen from each of the 1-2 intersection points; the step of reading a 2-1 road view screen from each of the 2-2 intersection points; and the step of reading a 3-1 road view screen from each of the 3-2 intersection points, a road view screen is read from each of the extracted intersection points.

[0117] In other words, 8 Road View 1-1 screens, 8 Road View 2-1 screens, and 8 Road View 3-1 screens are read.

[0118] In the step of extracting 2D bird's-eye view screens for each of the 8 directions of east, west, south, north, northeast, southeast, southwest, and northwest from the above bird's-eye view 3D modeling, 2D screens captured from the bird's-eye view 3D modeling are extracted by considering the direction in which the apartment will be placed when completed in the sales information.

[0119] In other words, 8 2D bird's-eye view screens are extracted by viewing the 3D bird's-eye view model from 8 different directions.

[0120] In the step of calculating a first orientation, which is the location of each of the first-2 intersection points based on the center of the first circle; the step of calculating a second orientation, which is the location of each of the second-2 intersection points based on the center of the first circle; and the step of calculating a third orientation, which is the location of each of the third-2 intersection points based on the center of the first circle; the first to third orientations are each calculated to determine which direction on the bird's-eye view is visible when looking at the apartment for sale (center of the first circle) from each intersection point.

[0121] In other words, 8 first directions, 8 second directions, and 8 third directions are produced.

[0122] In the step of generating a 1-2 road view screen by compositing the 2D bird's-eye view screen corresponding to the first orientation onto each of the 1-1 road view screens; the step of generating a 2-2 road view screen by compositing the 2D bird's-eye view screen corresponding to the second orientation onto each of the 2-1 road view screens; and the step of generating a 3-2 road view screen by compositing the 2D bird's-eye view screen corresponding to the third orientation onto each of the 3-1 road view screens, the 2D bird's-eye view screens corresponding to the first to third orientations are composited onto the 1-1 road view screen, the 2-1 road view screen, and the 3-1 road view screen.

[0123] At this time, based on an image object recognition model, a 'sky object' can be recognized from the 1-1 Road View screen, the 2-1 Road View screen, and the 3-1 Road View screen, and the 2D bird's-eye view screen can be superimposed and composited onto the center of the 'sky object'.

[0124] In the step of designating the above 1-1 Road View screens, 1-2 Road View screens, 2-1 Road View screens, 2-2 Road View screens, 3-1 Road View screens, and 3-2 Road View screens as expected bird's-eye view information; and the step of including the above expected bird's-eye view information in the above sales information; all 24 existing (actual) Road View screens read and 24 generated (synthesized) Road View screens are included in the sales information and returned.

[0125] Figures 4 (a) and 4 (c) show the current road view screen at two different points, and Figures 4 (b) and 4 (d) show the road view screen with a bird's-eye view superimposed at the corresponding points.

[0126] Additionally, the user terminal is controlled according to a control method further comprising: a step of displaying a second map on the user terminal; a step of marking points capable of displaying a bird's-eye view based on the expected bird's-eye view information on the second map; a step of reading a fourth road view screen for a selected bird's-eye view request point among the points capable of displaying a bird's-eye view from the expected bird's-eye view information; and a step of displaying the fourth road view screen; wherein the step of reading the fourth road view screen comprises: a step of reading a 4-1 road view screen, which is an existing road view corresponding to the bird's-eye view request point; and a step of reading a 4-2 road view screen, which is a composite road view corresponding to the bird's-eye view request point; and the step of displaying the expected bird's-eye view information comprises: a step of extracting a first image which is the difference between the 4-2 road view screen and the 4-1 road view screen; a step of extracting a first shape which is the outline of the first image; a step of generating a second image by converting the first image into grayscale; and a step of calculating a first pixel number which is the maximum horizontal pixel number of the first image. A step of calculating a second pixel number, which is the maximum vertical pixel number of the first image; a step of calculating a third pixel number, which is the horizontal pixel number of the 4-2 road view screen; 0.2% to 0.0% of the third pixel number.The method may include: a step of calculating a steel frame spacing of 3%; a step of calculating the number of steel frames, which is a value obtained by dividing the number of first pixels by the steel frame spacing; a step of calculating an inter-floor spacing, which is a value obtained by dividing the number of second pixels by 8 to 12; a step of displaying the 4-1 road view screen; a step of displaying the 1st shape on the upper layer of the 4-1 road view screen; a step of applying a glint animation to the 1st shape; a step of displaying multiple black vertical lines spaced apart from each other multiple times on the inner side of the 1st shape, wherein all black vertical lines corresponding to the number of steel frames are displayed within 3 to 4 seconds; a step of generating 3rd images by cutting the 2nd image from the bottom at each 'pixel corresponding to the inter-floor spacing'; a step of displaying the 3rd images by stacking them one by one from the bottom on the upper layer of the 1st shape; and a step of displaying the 1st image on the upper layer of the 3rd image.

[0127] In the step of displaying a second map on the above user terminal, the second map is displayed according to the operation of a button (bird's-eye view view button) on the user terminal.

[0128] In the step of marking points where a bird's-eye view can be displayed based on the predicted bird's-eye view information on the second map, the points where a bird's-eye view can be displayed (24 points) on the second map are highlighted and provided as buttons that the user can select.

[0129] In the step of reading a fourth road view screen for a selected bird's-eye view request point among the bird's-eye view displayable points from the above-mentioned expected bird's-eye view information, a road view (fourth road view screen; any one of 24 road views) for a bird's-eye view request point that is a point selected (touched, clicked) by the user on the user terminal among the above-mentioned 24 bird's-eye view displayable points is read.

[0130] In the step of displaying the fourth road view screen, the retrieved fourth road view screen is displayed (indicated) on the user terminal.

[0131] At this time, the user terminal can inspire anticipation in the user regarding the construction process of the apartment for sale based on a predetermined animation effect.

[0132] In the step of reading the 4-1 Road View screen, which is an existing Road View corresponding to the above-mentioned bird's-eye view request point; the existing Road View (4-1 Road View screen), which is one of a pair of Road Views included in the 4th Road View screen, is read. The 4-1 Road View screen may be any one of the 1-1 Road View screen, 2-1 Road View screen, and 3-1 Road View screen described above.

[0133] In the step of reading the 4-2 Road View screen, which is a composite Road View corresponding to the above-mentioned bird's-eye view request point; the other composite Road View (4-2 Road View screen) among the pair of Road Views included in the 4 Road View screen is read. The 4-2 Road View screen may be any one of the 1-2 Road View screen, 2-2 Road View screen, and 3-2 Road View screen described above.

[0134] In the step of extracting the first image which is the difference between the above 4-2 Road View screen and the 4-1 Road View screen; a portion of the first image which is the difference (pixels) between the actual Road View screen, the 4-1 Road View screen, and the composite Road View screen, the 4-2 Road View screen, is extracted.

[0135] In the step of extracting a first shape, which is the outline of the first image, only the outline of the first image (2D bird's-eye view portion) is extracted and designated as the first shape.

[0136] In the step of generating a second image by converting the first image into grayscale, a second image in black and white is generated separately based on the first image, which is a color view.

[0137] In the step of calculating the first pixel number, which is the maximum horizontal pixel number of the first image, the horizontal length of the first image (bird's-eye view) is calculated. In other words, the number of pixels of the part with the longest horizontal length among the parts constituting the first image is designated as the first pixel number.

[0138] In the step of calculating the second pixel number based on the longest vertical pixel number of the first image, the vertical length of the first image (bird's-eye view) is calculated. In other words, the number of pixels of the part with the longest vertical length among the parts constituting the first image is designated as the second pixel number.

[0139] In the step of calculating the third pixel number, which is the number of horizontal pixels of the above 4-2 Road View screen; not only the first image which is the bird's-eye view portion, but also the third pixel number, which is the total number of horizontal pixels of the 4-2 Road View screen which is the entire image including it, is calculated.

[0140] In the step of calculating a steel frame spacing that is 0.2% to 0.3% of the number of third pixels; and the step of calculating the number of steel frames that is the value obtained by dividing the number of first pixels by the steel frame spacing; the spacing and number of steel frames are calculated to display an image of a steel frame during the playback of the animation. In other words, the larger the 4-2 Road View screen, the more steel frames are displayed.

[0141] In the step of calculating the inter-floor spacing, which is a value obtained by dividing the second pixel count by 8 to 12, the inter-floor spacing is calculated to indicate the parts of the bird's-eye view cut vertically (parts depicting the 'floors' of the apartment) (this 'inter-floor spacing' value is unrelated to the actual number of floors of the apartment for sale).

[0142] The aforementioned steps are a preprocessing process performed before playback to make it appear as if a presale apartment is being constructed on the user's terminal.

[0143] According to the following steps, an animation showing an apartment for sale being constructed is played on the user terminal.

[0144] In the step of displaying the above 4-1 Road View screen; the 4-1 Road View screen, which is the actual current Road View screen, is displayed on the user terminal.

[0145] In the step of displaying the first shape on the upper layer of the above 4-1 Road View screen, the first shape is displayed by overlapping it on the upper layer of the above 4-1 Road View screen.

[0146] In other words, the first shape is displayed on one side of the 4-1 Road View screen, and since the interior of the first shape is transparent, the 4-1 Road View screen (sky portion) is displayed as is on the inside of the first shape.

[0147] In the step of applying a glint animation to the first shape, a sparkling effect is applied to the first shape to make the user aware of the overall shape of the apartment for sale.

[0148] In the step of marking multiple black vertical lines spaced apart from each other on the inner side of the first shape, and marking all black vertical lines corresponding to the number of steel frames within 3 to 4 seconds; black vertical lines are marked on the inner side of the first shape.

[0149] At this time, multiple (e.g., 7 to 9) black vertical lines are marked simultaneously, and this process is repeated for a short period of time to mark all black vertical lines corresponding to the number of steel frames (tens to hundreds).

[0150] Accordingly, black vertical lines are arranged at regular intervals inside the first shape, and this animation can provide the user with the impression that steel structures are being erected in the inner part of the first shape.

[0151] In the step of generating third images by cutting the second image from the bottom for each 'pixel corresponding to the interlayer interval', third images are generated by cutting the second image by the interlayer interval (height of one layer).

[0152] For example, if the second image is divided into 8 layers, 8 third images are generated.

[0153] In the step of stacking and displaying the third images one by one from the bottom on the upper layer of the first shape, the third images are stacked and displayed one by one from the bottom (from the first layer) on the inner side of the first shape.

[0154] At this time, the aforementioned black vertical lines are covered at the back of the third image and gradually become invisible.

[0155] Through these animations, users can be given the feeling that construction is in progress.

[0156] In the step of displaying the first image on the upper layer of the third image, by overlaying the first image again on the third image, which was a black and white (grayscale) image, the impression that painting (architecture) has been completed can be provided.

[0157] An apparatus according to one embodiment includes a processor and memory. The processor may include at least one apparatus described above through the drawings or perform at least one method described above through the drawings. The memory may store information related to the method described above or store a program in which the method described above is implemented. The memory may be volatile memory or non-volatile memory.

[0158] The processor can execute a program and control the device. The code of the program executed by the processor can be stored in memory. The device can be connected to an external device (e.g., a personal computer or a network) through an input / output device (not shown in the drawing) and exchange data.

[0159] The embodiments described above may be implemented as hardware components, software components, and / or combinations of hardware and software components. For example, the devices, methods, and components described in the embodiments may be implemented using one or more general-purpose or special-purpose computers, such as, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate array (FPGA), a programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions. The processing unit may execute an operating system (OS) and one or more software applications executed on said operating system. Additionally, the processing unit may access, store, manipulate, process, and generate data in response to the execution of the software. For ease of understanding, the processing unit may be described as being used as a single unit, but those skilled in the art will understand that the processing unit may include multiple processing elements and / or multiple types of processing elements. For example, the processing unit may include multiple processors or one processor and one controller. Additionally, other processing configurations, such as parallel processors, are also possible.

[0160] The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc., either alone or in combination. The program instructions recorded on the medium may be those specifically designed and configured for the embodiment, or they may be those known and available to those skilled in the art of computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes; optical recording media such as CD-ROMs and DVDs; magneto-optical media such as floptical disks; and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, and flash memory. Examples of program instructions include machine code, such as that generated by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc. The hardware devices described above may be configured to operate as one or more software modules to perform the operation of the embodiment, and vice versa.

[0161] Software may include computer programs, code, instructions, or a combination of one or more of these, and may configure a processing unit to operate as desired or instruct the processing unit independently or collectively. Software and / or data may be permanently or temporarily embodied in any type of machine, component, physical device, virtual equipment, computer storage medium, or device so as to be interpreted by the processing unit or to provide instructions or data to the processing unit. Software may be distributed over networked computer systems and stored or executed in a distributed manner. Software and data may be stored on one or more computer-readable recording media.

[0162] Although the embodiments have been described above with reference to the limited drawings, those skilled in the art can apply various technical modifications and variations based on the above. For example, suitable results may be achieved even if the described techniques are performed in a different order than described, and / or if the components of the described system, structure, device, circuit, etc. are combined or assembled in a form different from described, or replaced or substituted by other components or equivalents.

[0163] Therefore, other implementations, other embodiments, and equivalents to the claims also fall within the scope of the claims set forth below. Explanation of the symbols

[0164] S100: Step of receiving user information S200: Step for generating presale information S300: A step of transmitting sales information to the above user terminal C1: First element C2: Second element C3: Third element L1: First straight line L2: Second straight line L3: Third straight line L4: The 4th line A: Complex area 1 : Device (Server) 2 : User terminal 3 : Presale Expert Terminal 4 : Database

Claims

Claim 1 A method for providing a user-customized presale-related information provision and matching platform service utilizing an artificial intelligence model based on user status information analysis, wherein the device includes a processor, memory, a communication module, and a non-transient storage medium, and the program stored in the non-transient storage medium is executed by the processor, comprising: receiving user information from a user terminal; inputting the user information into a first artificial intelligence model to generate presale information; transmitting the presale information to the user terminal; matching presale expert information corresponding to the presale information to the user terminal; and transmitting the presale expert information to the user terminal. The method includes the step of activating a real-time consultation channel between a sales expert terminal corresponding to the above-mentioned sales expert information and the above-mentioned user terminal; wherein the above-mentioned user information includes: type of housing subscription account, housing subscription account subscription period, housing subscription account deposit amount, period of non-homeownership, family composition, number of dependents, income level, asset information, area of ​​interest, and housing preference conditions; and the above-mentioned sales information includes: sales complex information, priority, and subscription schedule; and the step of activating the above-mentioned real-time consultation channel includes: a first consultation step of inputting the above-mentioned user information into a second artificial intelligence model to generate first consultation assistance information; a second consultation step of transmitting the above-mentioned user information and the first consultation assistance information to the above-mentioned sales expert terminal; a third consultation step of adding conversation content additionally input into the above-mentioned real-time consultation channel to the second artificial intelligence model to generate second consultation assistance information; and a fourth consultation step of transmitting the second consultation assistance information to the above-mentioned sales expert terminal. and a fifth consultation step that repeats the third consultation step and the fourth consultation step until consultation termination information is received from the user terminal or the sales expert terminal; wherein the third consultation step comprises: a step of inputting the additionally input conversation content into the third artificial intelligence model to transform it into a standard conversation model; and a step of converting the standard conversation model into a first-1 vector;A step of generating a first-2 vector by multiplying the first-1 vector by a first weight; a step of reading existing conversation content corresponding to the user information; a step of converting the existing conversation content into a second-1 vector; a step of generating a second-2 vector by multiplying the second-1 vector by a second weight; a step of calculating a third-1 vector by summing the first-2 vector and the second-2 vector; a step of reading a first matrix from a database, wherein the first conversation example model is the 'row' and 'type of housing subscription account, housing subscription account subscription period, housing subscription account payment amount, period of non-homeownership, family composition, number of dependents, income level, asset information, area of ​​interest, housing preference conditions, and standard words' are the 'columns'; a step of generating a third-2 vector by adding the values ​​of 'type of housing subscription account, housing subscription account subscription period, housing subscription account payment amount, period of non-homeownership, family composition, number of dependents, income level, asset information, area of ​​interest, and housing preference conditions' read from the user information to the third-1 vector; The method comprises: a step of generating a second matrix by adding a third-2 vector to the first matrix; a step of classifying the second matrix into first clusters based on a K-Means Clustering model; a step of extracting a second cluster among the first clusters that is a first cluster containing the first-1 vector; a step of receiving an existing conversation proposal list from the sales expert terminal; a step of extracting a second conversation example model among the 'rows' included in the second cluster that do not correspond to the 'third-2 vector and the existing conversation proposal list'; and a step of designating the second conversation example model as the second consultation assistance information; wherein the user terminal comprises: a step of receiving a rectangle on a first map surrounding the area of ​​interest; and a step of designating the rectangle as the area of ​​interest.Controlled according to a control method including, the step of generating the 3-2 vector comprises: receiving the coordinates of each vertex of the rectangle constituting the region of interest, (x11, y11), (x11, y12), (x12, y11), and (x12, y12) (where x11 < x12 and y11 < y12); and substituting x3 = (x12 - x11) / 5 and y3 = (y12 - y11) / 5; A step of substituting x21 = x11, x22 = x11 + x3, x23 = x11 + 2*x3, x24 = x11 + 3*x3, x25 = x11 + 4*x3, x26 = x12, y21 = y11, y22 = y11 + y3, y23 = y11 + 2*y3, y24 = y11 + 3*y3, y25 = y11 + 4*y3, and y26 = y12; a step of calculating first coordinates, which are 36 grid point coordinates formed based on x21 to x26 and y21 to y26; and the step of assigning the 36 first coordinates to the 36 pre-designated 'column' positions among the 'columns' of the 3-2 vector; and the step of generating the sales information comprises: the step of reading the complex address from the sales complex information; the step of reading the complex area information from the sales complex information; the step of reading the bird's-eye view 3D model from the sales complex information; the step of drawing a first circle passing outside the complex on a first map based on the complex address and complex area information; the step of drawing a second circle on the first map that is concentric with the first circle and has a diameter of 105% to 108% of the diameter of the first circle; the step of drawing a third circle on the first map that is concentric with the first circle and has a diameter of 115% to 120% of the diameter of the first circle; and the step of drawing a first straight line, which is a vertical line passing through the center of the first circle, on the first map. A step of drawing a second straight line, which is a horizontal line passing through the center of the first circle, on the first map;A step of drawing a third line and a fourth line on a first map, which are diagonals passing through the center of the first circle and forming a 45-degree angle with the vertical and horizontal lines; a step of extracting 1-1 intersection points, which are intersection points between the first circle and the first to fourth lines; a step of extracting 2-1 intersection points, which are intersection points between the second circle and the first to fourth lines; a step of extracting 3-1 intersection points, which are intersection points between the third circle and the first to fourth lines; a step of extracting 1-2 intersection points, which are points on the first to fourth lines that are closest to each of the 1-1 intersection points among points on the first map where a road view exists; a step of extracting 2-2 intersection points, which are points on the first to fourth lines that are closest to each of the 2-1 intersection points among points on the first map where a road view exists. Among the points where a road view exists on the first map, a step of extracting third-2 intersection points that are closest to each of the third-1 intersection points and are points on the first to fourth straight lines; a step of reading a first-1 road view screen from each of the first-2 intersection points; a step of reading a second-1 road view screen from each of the second-2 intersection points; a step of reading a third-1 road view screen from each of the third-2 intersection points; a step of extracting a 2D bird's-eye view screen for each of the eight directions of east, west, south, north, northeast, southeast, southwest, and northwest from the bird's-eye view 3D modeling; a step of calculating a first direction, which is the location of each of the first-2 intersection points, based on the center of the first circle; a step of calculating a second direction, which is the location of each of the second-2 intersection points, based on the center of the first circle; a step of calculating a third direction, which is the location of each of the third-2 intersection points, based on the center of the first circle; A step of generating a 1-2 road view screen by compositing the 2D bird's-eye view screen corresponding to the 1st orientation onto each of the 1-1 road view screens;A method for providing a user-customized presale-related information provision and matching platform service utilizing an artificial intelligence model based on user status information analysis, comprising: a step of generating a 2-2 Roadview screen by synthesizing the 2D bird's-eye view screen corresponding to the second orientation onto each of the 2-1 Roadview screens; a step of generating a 3-2 Roadview screen by synthesizing the 2D bird's-eye view screen corresponding to the third orientation onto each of the 3-1 Roadview screens; a step of designating the 1-1 Roadview screens, 1-2 Roadview screens, 2-1 Roadview screens, 2-2 Roadview screens, 3-1 Roadview screens, and 3-2 Roadview screens as expected bird's-eye view information; and a step of including the expected bird's-eye view information in the presale information; Claim 2 delete Claim 3 delete

Citation Information

Patent Citations

  • Apparatus and method for providing of trading area information

    KR101816215B1

  • Method, apparatus and system for providing agency service of selling of real estate

    KR1020200025985A

  • Apparatus, system, method and program for providing presale marketing support service

    KR102524647B1

  • Method and device for recommending real estate for sale using artificial intelligence-based conversational chatbot and a computer-readable recording medium recording a program for executing the method

    KR102656643B1