KR-102962719-B1 - Prediction system for whether a real estate rights holder will exercise a put option using AI big data learning

KR102962719B1KR 102962719 B1KR102962719 B1KR 102962719B1KR-102962719-B1

Abstract

The present invention is characterized by comprising: a data collection unit that defines pre-set big data related to whether a real estate right holder will exercise a put option and selectively acquires and collects the defined pre-set big data; an AI learning unit that learns the pre-set big data selectively acquired and collected by the data collection unit; and an exercise probability evaluation unit that calculates the probability of the real estate right holder exercising the put option as a quantitative value based on the pre-set big data learned by the AI learning unit.

Inventors

김종구

Assignees

한국자산매입 주식회사

Dates

Publication Date: 20260508
Application Date: 20250704

Claims (10)

In a system for predicting whether a real estate rights holder will exercise a put option by utilizing AI (Artificial Intelligence) and collecting and learning big data, A data collection unit that defines pre-set big data related to whether the aforementioned real estate right holder will exercise a put option, and selectively acquires and collects the defined pre-set big data; An AI learning unit that selectively acquires and collects the above-mentioned data collection unit and performs AI learning on the above-mentioned preset big data; An exercise probability evaluation unit that calculates the probability of the real estate right holder exercising a put option as a quantitative value based on the preset big data learned by the AI learning unit; and It includes an exercise probability security processing unit that securely processes information regarding the put option exercise probability, which is information regarding the calculated exercise probability, The above event probability security processing unit is, An exercise probability information block processing unit that represents information on the exercise probability of the above put option as a binary signal of 1 and 0, arranges the binary signal into a binary array, groups the binary array into blocks according to a predetermined number of binary codes, and assigns a unique address to each of the blocks; A block order reordering unit that reorders by applying a function rule to reorder the position of each block unit of the above-mentioned event probability information block processing unit; and A system for predicting whether a real estate right holder will exercise a put option using AI big data learning, characterized by providing a stack capable of stacking a set of block units rearranged by the block order rearrangement unit, and including a block stack processing unit that sequentially stacks N rearranged blocks in each of the stacks.
In paragraph 1, the above system is, The above event probability, A prediction system for whether a real estate right holder will exercise a put option using AI big data learning, characterized by defining the exercise of a purchase claim right regarding the right held by the real estate right holder or the real estate registration right holder in relation to the said real estate.
In paragraph 2, the data collection unit is, A customer information collection unit that sets the above-mentioned real estate rights holder as a customer and selectively collects and acquires customer information set as part of the above-mentioned pre-set big data; and A system for predicting whether a real estate right holder will exercise a put option using AI big data learning, characterized by including a transaction trend information collection unit that sets a preset boundary for the real estate in which the real estate right holder holds rights, sets preset transaction trend information for real estate within the preset boundary, and selectively collects and acquires the preset transaction trend information as part of the preset big data.
In paragraph 3, the data collection unit is, A system for predicting whether a real estate right holder will exercise a put option using AI big data learning, characterized by defining domestic economic indicators related to the real estate mentioned above and further including a domestic variable collection unit that selectively collects and acquires the domestic economic indicators mentioned above.
In paragraph 4, the data collection unit is, A prediction system for whether a real estate right holder will exercise a put option using AI big data learning, characterized by further including an external variable collection unit that defines external economic indicators related to the real estate and selectively collects and acquires the external economic indicators.
In Paragraph 3, the customer information collection unit above, As the above-mentioned preset customer information, the customer's, A system for predicting whether a real estate right holder will exercise a put option using AI big data learning, characterized by including at least one of i) financial factor information, ii) real estate ownership history information, iii) information on whether the purpose is actual residence, or iv) social/behavioral pattern information.
In paragraph 3, the above transaction trend information collection department is, If the above property is an apartment, the above preset boundary is defined as the apartment complex of the above property, and A system for predicting whether a real estate rights holder will exercise a put option using AI big data learning, characterized by including, as the above-mentioned preset transaction trend information, at least one of i) actual transaction price trend information of the same type of apartment within the administrative district to which the above-mentioned apartment complex belongs, ii) transaction volume information of the same type of apartment, iii) rental income information of the same type of apartment, or iv) specific information of the above-mentioned apartment complex.
In paragraph 4, the above internal variable collection unit If the above property is an apartment, the above preset boundary is defined as the apartment complex of the above property, and A system for predicting whether a real estate rights holder will exercise a put option using AI big data learning, characterized by including, as the above-mentioned domestic economic indicators, at least one of i) brand information of the above-mentioned apartment complex, ii) information on the conditions of the presale contract of the above-mentioned apartment complex, iii) information on subsequent changes of the above-mentioned apartment complex, or iv) information on the contract trends of presale buyers of the above-mentioned apartment complex.
In paragraph 5, the above-mentioned event probability evaluation unit is, A prediction system for whether a real estate right holder will exercise a put option using AI big data learning, characterized by including a purchase claim exercise probability calculation unit that calculates the probability of the real estate right holder exercising a put option as a quantitative value based on the pre-set big data learned by the AI learning unit based on the apartment complex to which the real estate belongs, when the real estate is an apartment.
In Clause 9, the above-mentioned event probability evaluation unit is, A prediction system for whether a real estate right holder will exercise a put option using AI big data learning, characterized by further including a portfolio exercise probability calculation unit that quantitatively calculates the average probability of the real estate right holders exercising a put option for the entire apartment complex of the portfolio composed of multiple portfolios, based on the preset big data learned by the AI learning unit, when the real estate is an apartment and the apartment complex to which the real estate belongs constitutes multiple portfolios.

Description

Prediction system for whether a real estate rights holder will exercise a put option using AI big data learning The present invention relates to a system for predicting whether a rights holder, such as a real estate owner or a holder of a presale right, will exercise the right to sell a property they own or have prescribing. More specifically, the invention relates to a technical field concerning a system that utilizes artificial intelligence (AI) and collects and learns from big data to predict the probability of a person holding rights to real estate, such as an apartment, exercising a purchase right when reaching a time agreed upon for exercising the purchase right, either after finally purchasing the said apartment or without purchasing it. In the apartment presale market, whether pre-purchasers fully pay their subscription fees is directly linked to the construction company's capital recovery and the stability of project finance. Delays or non-payments increase the developer's risk of bankruptcy, which leads to an increase in non-performing loans for financial institutions and places a burden on the overall financial system. Furthermore, it can cause social unrest through contract cancellations by investors based on pre-sale rights, an increase in lawsuits, and delays in occupancy for actual buyers, potentially triggering ripple effects that lead to a decline in confidence across the entire housing market. Transactions involving pre-sale rights or properties scheduled for presale are gradually evolving from a past offline-centric model to real-time trading platforms utilizing computerized systems. These platforms resolve information asymmetry between buyers and sellers, enhance the safety and transparency of contract execution, and foster trust among market participants. In particular, AI-based price prediction, real-time property matching, and blockchain-based contract management complement the limitations of the existing real estate brokerage market and possess high market potential and growth possibilities, as they can dramatically improve transaction efficiency in the large-scale pre-sale market. "A method for auctioning unsold apartments using the Internet (Publication No. 10-2001-0091222, Patent Document 1)" exists. The invention of Patent Document 1 provides an internet auction method comprising the steps of: a plurality of apartment construction companies accessing an auction server having a database and storing information on a plurality of unsold apartments, including a minimum sale price and an auction end date, in the database; a user who wishes to purchase an unsold apartment accessing the auction server, registering as a member, and searching for the stored information on a plurality of unsold apartments; a member applying for a bid on a specific unsold apartment among the information on a plurality of unsold apartments at a price equal to or greater than the minimum sale price; and selecting a successful bidder among the applicants who applied before the auction end date. "An electronic trading system for apartment presale rights using the Internet (Publication No. 10-2002-0022439, Patent Document 2)" exists. The invention of Patent Document 2 relates to an electronic transaction system for apartment presale rights using the Internet, which specializes in electronic transactions for apartment presale rights on the Internet to enable users to view all information regarding presale rights with a single click from home, and ensures the reliability of presale right transactions by having licensed real estate agents mediate the presale rights. The invention of Patent Document 2 relates to an electronic transaction system for apartment presale rights using the Internet, comprising: a buyer terminal requesting member registration and the purchase of an apartment presale right; a seller terminal requesting member registration and the sale of an apartment presale right; a server connected to the buyer terminal or the seller terminal via an Internet communication network, which receives member registration and requests for the purchase and sale of apartment presale rights in accordance with the terminal's request to induce electronic transactions, and provides various information regarding the presale right; a member DB storing member information requested for registration on the server; and a presale right listing DB storing presale right sales information requested for registration on the server. It includes a presale right purchase application DB that stores presale right purchase information requested for registration on the server, and a presale right information DB that stores financial information, tax information, legal information, presale information, and presale right Q&A information provided by the server. FIG. 1 is a conceptual diagram illustrating the relationship between a real estate rights holder and a construction developer using AI big data learning to predict whether a put opti