KR102293329B1 - Method and apparatus for providing real estate information - Google Patents

Abstract

The present invention relates to a method and an apparatus for providing real estate information. According to an embodiment of the present invention, the method comprises the steps of: collecting real estate sales information and information about a real estate consumer including the tendency information of the real estate consumer; forming a relationship table between the real estate consumer and the real estate sale based on the information on the real estate consumer and the real estate sale information, wherein the relationship table includes a weight value for each real estate sale calculated according to the tendency of the real estate consumer; analyzing the similarity between the real estate consumers using weights for each real estate sale for the real estate consumer in the relationship table; and providing the real estate sales suitable for the real estate consumer in the relation table or at least one new real estate consumer based on a similarity analysis result between the real estate consumers.

Description

Method and apparatus for providing real estate information

The present invention relates to real estate information providing technology, and more particularly, to a method and apparatus for providing real estate information customized to real estate consumers.

In the traditional form of real estate transactions that have been carried out in the past, real estate buyers (eg, real estate buyers, real estate tenants, etc.) contract with the seller, real estate lessor, etc.). Therefore, according to the traditional real estate transaction form, the real estate consumer had to endure the inconvenience or inconvenience of having to visit real estate agencies in various desired areas one by one.

In today's world where Internet technology is common, real estate sales information that could be obtained through real estate agencies in the past is also provided on Internet websites. For example, the real estate brokerage company provides information on the real estate for sale that it owns through Internet websites or real estate information providers (Naver Real Estate, Daum Real Estate, Real Estate 114, Jikbang, Coffee Shop, etc.) It is posted on the Internet website. Accordingly, a recent real estate consumer can conveniently search for a real estate for sale by accessing the Internet website without restriction of place or time.

In the conventional real estate sale recommendation method, real estate information such as the price, location, and acreage of the real estate for sale posted as a transaction by the real estate provider is stored in a database, and the real estate receiver stores his/her budget, desired area, When information such as the number of square feet is input, real estate corresponding to the input information is displayed or recommended as a list.

However, factors considered in the conventional real estate sale recommendation are very limited to information such as price, desired area, number of acres, and option existence, and other real estate properties that have a possibility of being suitable for the real estate buyer other than the search condition of the real estate buyer are recommended Not only cannot receive it, but it cannot be recommended at all for a property that has an excellent suitability for an area that the real estate buyer did not think of.

The technical problem to be achieved by the present invention is to provide a method of providing improved real estate information for recommending real estate for sale that is likely to be suitable for the real estate consumer, without being limited to information considered for recommending real estate for sale.

In addition, another technical problem to be achieved by the present invention is to provide an apparatus for providing improved real estate information having the above-described advantages.

According to an embodiment of the present invention, there is provided an apparatus for providing real estate information, comprising: one or more processors; and a memory, wherein the processor collects and stores information about the real estate consumer and real estate sale information including the propensity information of the real estate consumer, and stores it in the memory, based on the information about the real estate consumer and the real estate sale information , to form a relationship table between the real estate consumer and the real estate sale, the relationship table includes a weight value for each real estate sale calculated by the propensity of the real estate consumer, and a weight for each real estate sale for the real estate consumer in the relationship table It is possible to analyze the similarity between the real estate consumers using the method, and control to provide a suitable real estate sale to the real estate consumer or at least one new real estate consumer in the relationship table based on the similarity analysis result between the real estate consumers. The similarity between the real estate consumers is a Pearson correlation coefficient, cosine similarity, adjusted cosine similarity, mean squared difference similarity, jaccard similarity or It can be determined from a combination thereof. The propensity information of the real estate consumer may be determined from the personal details that can be viewed and provided based on the membership registration information of the real estate information providing service, the usage pattern of the real estate information providing service, the search pattern or authentication of related real estate information, or a combination thereof. can

In one embodiment, real estate consumers in the relationship table can be subdivided through the personal details, form a group with a similar tendency, and real estate buyers who form a group with a similar tendency are interested in real estate listings in the relationship table It may include real estate listings that have

Weight values not calculated in the relation table are predicted through a supervised learning algorithm, and the supervised learning algorithm is K-Nearest Neighbors, Linear Regression, Logistic Regression, It may include any one of a support vector machine (SVM), a decision tree, a random forest, and an artificial neural network.

In an embodiment, the processor may update the weight of the real estate for sale in the relationship table by reflecting a review or rating for at least one real estate for sale among current residents and other real estate buyers of the real estate for sale. The processor may predict the future real estate price of the real estate for sale through a learned artificial intelligence engine, and provide information on the future real estate price of the predicted real estate for sale to the real estate consumer. The real estate for sale may include any one of a first real estate that is currently for sale and a second real estate that is likely to be for sale within the next several months.

In one embodiment, the information on the real estate consumer includes profile information of the real estate consumer required at the time of membership registration and detailed personal information of the real estate consumer, and the tendency information of the real estate consumer is the search history of the real estate consumer, related real estate information A building that is determined through the history of use of the provided service or a combination thereof, and the real estate sale information includes information on the location, desired transaction type, expected sale period, acreage, management fee, housing type, housing direction, internal structure, completion date, and price information traffic safety, traffic accessibility, convenience facilities, security facilities, medical facilities, schools and academies, sex offenders' residences, and surrounding environment-related information such as areas where policing accidents are common. The detailed personal information of the real estate consumer may include personal information that can be viewed and provided through an authentication means such as property, marital status, number of family members, wages, and card details.

According to another embodiment of the present invention, collecting information and real estate sales information about the real estate consumer, including the tendency information of the real estate consumer; Forming a relationship table between a real estate consumer and a real estate sale based on the information on the real estate consumer and the real estate sale information, wherein the relationship table includes a weight value for each real estate sale calculated by the propensity of the real estate consumer; analyzing the similarity between real estate consumers using weights for each real estate sale for the real estate consumers in the relationship table; There may be provided a real estate information providing method comprising providing a real estate for sale suitable for a real estate consumer in the relation table or at least one new real estate consumer based on a result of the analysis of the similarity between the real estate consumers. The similarity between the real estate consumers is a Pearson correlation coefficient, cosine similarity, adjusted cosine similarity, mean squared difference similarity, jaccard similarity or It can be determined from a combination thereof. The propensity information of the real estate consumer may be determined from the personal details that can be viewed and provided based on the membership registration information of the real estate information providing service, the usage pattern of the real estate information providing service, the search pattern or authentication of related real estate information, or a combination thereof. can Real estate buyers in the relationship table can be subdivided through the personal details, form a group with a similar tendency, and real estate sales in the relationship table are interested in real estate buyers who form a group with a similar tendency may include.

Weight values not calculated in the relation table are predicted through a supervised learning algorithm, and the supervised learning algorithm is K-Nearest Neighbors, Linear Regression, Logistic Regression, It may include any one of a support vector machine (SVM), a decision tree, a random forest, and an artificial neural network. The step of updating the weight of the real estate for sale in the relation table by reflecting the reviews or ratings for the real estate sale of the real estate consumer and the current resident of the real estate sale may be further included. In addition, predicting the future real estate price of the real estate sale through the learned artificial intelligence engine; And the step of providing information about the future real estate price of the predicted real estate for sale to the real estate consumer may be further included. The real estate for sale may include any one of a first real estate that is currently for sale and a second real estate that is likely to be for sale within the next several months.

According to an embodiment of the present invention, a relationship table is formed between a real estate buyer and a real estate sale including a weight value for each real estate sale calculated by the propensity of the real estate consumer, and a weight value for each real estate sale for the real estate consumer in the relationship table By analyzing the similarity between real estate buyers using A method for providing real estate information may be provided.

Further, according to another embodiment of the present invention, an apparatus for providing improved real estate information having the above advantages may be provided.

1 is a block diagram of a system for providing real estate information according to an embodiment of the present invention.
2 is a table showing a relationship between a real estate consumer and a real estate sale according to an embodiment of the present invention.
3A and 3B are diagrams illustrating an example of recommending a real estate sale to a real estate consumer according to an embodiment of the present invention.
4 is a flowchart illustrating a method for providing real estate information according to an embodiment of the present invention.

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

Examples of the present invention are provided to more completely explain the present invention to those of ordinary skill in the art, and the following examples may be modified in various other forms, and the scope of the present invention is as follows It is not limited to an Example. Rather, these examples are provided so as to more fully and complete the present invention, and to fully convey the spirit of the present invention to those skilled in the art.

Also, as used herein, the terms “component,” “module,” “system,” and the like refer to a computer-related entity, hardware, firmware, software, a combination of software and hardware, or execution of software. For example, a component can be, but is not limited to being, a process running on a processor, a processor, an item, a thread of execution, a program, and/or a computer. For example, both an application running on a computing device and the computing device may be a component. One or more components may reside within a processor and/or thread of execution. A component may be localized within one computer. A component may be distributed between two or more computers. In addition, these components can execute from various computer readable media having various data structures stored therein. Components may communicate via a network such as the Internet with another system, for example via a signal having one or more data packets (eg, data and/or signals from one component interacting with another component in a local system, distributed system, etc.) may communicate via local and/or remote processes depending on the data being transmitted).

The terminology used herein is used to describe specific embodiments, not to limit the present invention. As used herein, the singular forms may include the plural forms unless the context clearly dictates otherwise. Also, as used herein, “comprise” and/or “comprising” refers to the presence of the recited shapes, numbers, steps, actions, members, elements, and/or groups of those specified. and does not exclude the presence or addition of one or more other shapes, numbers, movements, members, elements and/or groups.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention are described with reference to the drawings which schematically show ideal embodiments of the present invention. In the drawings, for example, the size and shape of the members may be exaggerated for convenience and clarity of description, and in actual implementation, variations of the illustrated shape may be expected. Accordingly, embodiments of the present invention should not be construed as limited to the specific shapes of the members or regions shown herein. Hereinafter, various embodiments of the present invention will be described with reference to the drawings.

1 is a block diagram of a system for providing real estate information according to an embodiment of the present invention.

Referring to FIG. 1 , the system 100 accesses the real estate information providing device 10 including a processor 11 and a memory 12 and the real estate information providing device 10 through a communication network CN to access the corresponding real estate information. It may include a plurality of electronic devices (MS _{_1} , MS _{_2} , ..., MS _{_m} ) capable of obtaining . The configuration of the real estate information providing apparatus 10 shown in FIG. 1 is only a simplified example. In an embodiment of the present invention, the real estate information providing device 10 is a server that provides real estate information managed and operated by real estate brokerage firms or real estate information providers, and other components may be further included. For example, the real estate information providing apparatus 10 may further include an input/output interface module, a display, and a communication interface module.

The processor 11 may include one or more of a central processing unit (CPU), an application processor (AP), and a communication processor (CP). The processor 11 may, for example, execute operations or data processing related to control and/or communication of at least one other component of the device 10 .

In an embodiment of the present invention, the processor 11 uses a collaborative filtering-based algorithm to determine a real estate for sale that is likely to be suitable for the real estate consumer, and displays the result to the electronic device ( You can recommend real estate for sale through MS _{_1} , MS _{_2} ,…,MS _{_m ).} Specifically, with reference to the table of FIG. 2 to be described later, the processor 11 forms a relationship table between the real estate consumer and the real estate for sale by utilizing real estate-related information stored in the memory 12 to be described later, and then shows the tendency of the real estate consumer. Based on this, a weight value (or scoring or rating) for each real estate sale can be calculated. Through this, the similarity between real estate buyers can be analyzed and compared to recommend a property suitable for new real estate buyers. In addition, the processor 11 ) can determine the predicted value using a supervised learning-based algorithm for the weight value (or scoring) between the real estate buyer and the sale that is not in the relationship table, and thereby can recommend a customized real estate sale to the real estate consumer. The processor 11 is also capable of recommending a suitable property based on the existing learning information even if the tendency of the real estate consumer changes.

Memory 12 may include volatile and/or non-volatile memory. Memory 12 may store, for example, instructions or data related to at least one other component of device 10 . According to one embodiment, memory 12 may store software and/or programs. The program may include, for example, a kernel, middleware, an application programming interface (API), and/or an application program (or "application"). At least a part of the kernel, the middleware, or the API may be referred to as an operating system (OS). The kernel includes, for example, system resources (eg, processor 11 , or memory 12 ) used to execute operations or functions implemented in other programs (eg, middleware, API, or application program). )) can be controlled or managed. In addition, the kernel may provide an interface capable of controlling or managing system resources by accessing individual components of the device 10 from the middleware, the API, or the application program. The middleware may, for example, perform an intermediary role so that the API or the application program communicates with the kernel to exchange data. Also, the middleware may process one or more work requests received from the application program according to priority. For example, the middleware may give priority to use the system resource of the device 11 to at least one of the application programs. For example, the middleware) may perform a function such as scheduling or load balancing for the one or more work requests by processing the one or more work requests according to the priority given to the at least one. The API is, for example, an interface for the application to control a function provided by the kernel or the middleware, for example, at least one interface for file control, window control, image processing, or character control. Alternatively, it can contain functions (eg, commands).

In an embodiment of the present invention, the memory 12 may store real estate consumer-related information, real estate sale-related information, and detailed personal information of the real estate consumer as real estate-related information. The real estate consumer-related information is information required at the time of membership registration, and may include profile information of the real estate consumer such as name, gender, age, current residence, desired price range, desired area, number of square feet, or number of rooms. The real estate consumer information may further include propensity information of the real estate consumer that can be inferred by the search pattern and behavior pattern of the real estate consumer. Specifically, the propensity information of the real estate consumer may be analyzed through the real estate consumer's Internet search history, related real estate information providing service use history, or a combination thereof. The above real estate sale information includes location, desired transaction type (sale/jeonse/monthly rent), expected sale period, square footage, management fee, housing type (apartment, villa, country house, etc.), housing direction, internal structure, completion date, and price information. Detailed information on buildings, including traffic safety (location of frequent traffic accident sites, child accident sites, etc.), transportation accessibility (KTX, subway, bus stop locations, etc.), convenience facilities (department store, market, mart, park location, etc.), It may include information related to the surrounding environment, such as public security facilities (police stations, CCTV locations, etc.), medical facilities (hospital locations, etc.), schools and academies. The detailed personal information of the real estate consumer may be defined as personal information that can be viewed and provided through an authentication means such as property, marital status, number of family members, wages, and card details. The authentication means may include public authentication or authentication through a mobile phone or e-mail that can confirm whether it is an individual.

The input/output interface (not shown) may serve as an interface capable of transmitting, for example, a command or data input from a user or other external device to other component(s) of the device 10 . In addition, the input/output interface may output a command or data received from other component(s) of the device 11 to a user or other external device.

A display not shown may include, for example, a liquid crystal display (LCD), a light emitting diode (LED) display, an organic light-emitting diode (OLED) display, or a microelectromechanical display. microelectromechanical systems (MEMS) displays, or electronic paper displays. The display may, for example, display real estate related information to the user. The display may include a touch screen, and may receive, for example, a touch, gesture, proximity, or hovering input using an electronic pen or a part of the user's body.

The communication interface not shown is, for example, the electronic device 101 and an external device (eg, the smart garment 180 , the first external electronic device 102 , the second external electronic device 104 , or the server 106 ). )) can establish communication between them. For example, the communication interface 170 may be connected to the network 162 through wireless communication or wired communication to communicate with an external device (eg, the second external electronic device 104 or the server 106 ).

The communication network CN may include a wired or wireless communication network or a combination thereof. The wireless communication network is a cellular communication network, for example, long-term evolution (LTE), LTE Advance (LTE-A), code division multiple access (CDMA), wideband CDMA (WCDMA), universal mobile telecommunications system (UMTS), At least one of Wireless Broadband (WiBro) or Global System for Mobile Communications (GSM) may be used. In addition, the wireless communication may include, for example, short-range communication, wherein the short-distance communication is wireless fidelity (WiFi), Bluetooth, near field communication (NFC), infrared communication, Zigbee communication, and ultra-low-range communication (UWB). -wideband) communication, RFID (Radio-Frequency Identification), WiFi communication, or GNSS (global navigation satellite system) may include at least one of. However, the present invention is not limited thereto, and may include 4G and 5G networks.

In an embodiment, the electronic devices MS _{_1} , MS _{_2} , ..., MS _{_m} are electronic devices used by real estate buyers, and may search real estate information and display real estate for sale recommended by the real estate information providing device 10 . . The electronic devices MS _{_1} , MS _{_2} , ..., MS _{_m} may include any one of a smart phone, a smart pad, a smart note, a desktop, and a laptop. However, the present invention is not limited thereto.

2 is a table showing a relationship between a real estate consumer and a real estate sale according to an embodiment of the present invention. 2 illustrates a relationship table between 7 real estate buyers and 9 real estate listings, the present invention can be extended to a relationship table of size M * N. The size of the relation table may be determined by the total real estate consumer information and the total real estate sale information stored in the database, and may be subdivided according to the type of sale (apartment, villa, single/multi-family).

2, the weight values (or scoring, rating) between the first real estate receiver and the real estate sale 1 to the real estate sale 9 are 1, 3, 9, 9, 2, 3, 8, 4, 3, and the weight values (or scoring, rating) between the second real estate receiver and the real estate sale 1 to the real estate sale 9 are 3, 5, 7, 7, 8, 5, ?, 2, respectively , 8, and the weight values (or scoring, rating) between the third real estate receiver and the real estate sale 1 to the real estate sale 9 are 3, 3, 2, ?, 3, 6, 9, 7, 1, and the weight values (or scoring, rating) between the fourth real estate receiver and the real estate sale 1 to the real estate sale 9 are 2, 3, 3, 2, 8, 8, 3, 6, 4, respectively and the weight values (or scoring, rating) between the fifth real estate receiver and the real estate sale 1 to the real estate sale 9 are 2, 3, 3, 2, 8, 8, 3, 6, 4, respectively , the weight values (or scoring, rating) between the sixth real estate receiver and the real estate sale 1 to the real estate sale 9 are 5, 3, 8, 2, 8, 1, 3, 8, 9, respectively, The weight values (or scoring, rating) between the seventh real estate receiver and the real estate sale 1 to the real estate sale 9 are 1, 4, 5, 3, 1, 3, 4, 5, 6, respectively.

A weight value between the real estate consumer and the real estate for sale may be determined based on the propensity of the real estate consumer. Specifically, the tendency of the real estate consumer may be determined based on the real estate consumer-related information stored in the memory 12 and the detailed personal information of the real estate consumer, and a weight value may be determined according to the real estate consumer's tendency. Specifically, in the relation table between the real estate consumer and the real estate for sale, the weight may be determined as a score calculated for the sale evaluated based on the preference information previously selected by the user as shown in Equation 1 below.

<Equation 1>

은 최종 점수이고,

은 매물 기초 정보이고,

은 사용자 선호 정보이다. here,

is the final score,

is basic information for sale,

is user preference information.

In an embodiment, the basic information for sale may include absolute evaluation items for sale, such as building evaluation (complex size, age of retirement), management costs, and nearby convenience facilities. In addition, the user preference information may include user preference items such as school districts and sale prices.

In an embodiment, the degree of similarity between the real estate consumers may be determined through the relationship table of FIG. 2 . The similarity between the real estate consumers is a Pearson correlation coefficient, cosine similarity, adjusted cosine similarity, mean squared difference similarity, jaccard similarity, or combinations thereof may be included. By analyzing and comparing the similarity between the real estate buyers, it is possible to recommend real estate properties suitable for new real estate buyers.

In an embodiment, the degree of similarity by the Pearson correlation coefficient may be defined by the following <Equation 2>.

<Equation 2>

u _a , u _i are the real estate consumer indexes, w _p (u _a , u _i ) is the Pearson correlation coefficient between _{u a} and u _{i , and X aj} is the rating of the real estate buyer u _{a for the real estate sale j} , n _a is the average rating of real estate for sale by real estate buyer u _{a , X ij is the rating of real estate buyer u i} for real estate sale j, and n _i is the average rating of real estate for sale by real estate buyer u _i.

In an embodiment, cosine similarity may be defined by Equation 3 below.

<Equation 3>

u _a , u _i are the real estate buyer indexes, w _c (u _a , u _i ) is the cosine similarity value between _{u a} and u _{i , X aj} is the rating of the real estate buyer u _{a for the real estate sale j} , X _ij is the rating of _{the real estate consumer u i} for the real estate sale j.

In an embodiment, the weight value may exist in the relationship table in a state in which it is not calculated. This may occur when a new real estate buyer does not input additional points for evaluation of user preference information. For example, in FIG. 2, the weight is in an undetermined state (?) between the second real estate consumer and the real estate sale 7, and the weight is in an undetermined state (?) between the third real estate consumer and the real estate sale 4, and the fifth real estate buyer and The weight is undecided (?) among the real estate for sale 5.

In this case, the weight value that is not calculated in the relation table may predict an additional point for evaluating user preference information through a supervised learning algorithm. The supervised learning algorithm includes K-Nearest Neighbors, Linear Regression, Logistic Regression, Support Vector Machine (SVM), Decision Tree and Random It may include any one of a forest (Random Forest) and an artificial neural network (Artificial Neural Network). The artificial neural network may include machine learning and deep learning. The present invention is not limited to these algorithms.

In one embodiment, in the real estate sale recommendation model, based on the recommended predicted sale and feedback information (eg, a review or rating for the corresponding real estate sale) of the user (eg, current resident or other real estate buyers), The weight value can be updated. In addition, the update of these weight values may be learned through machine learning and deep learning.

3A and 3B are diagrams illustrating an example of recommending a real estate sale to a real estate consumer according to an embodiment of the present invention. 3A is a case of recommending real estate for sale based on user-based collaborative filtering, and FIG. 3B is a case of recommending real estate for sale based on item-based collaborative filtering. The similarity between users and the similarity between items may be determined from the relationship table of FIG. 2 described above. The user-based collaborative filtering is advantageous when the amount of data is small and changes to preferences occur frequently, and the item-based collaborative filtering is advantageous when the amount of data is large and changes in preferences do not occur frequently.

Referring to FIG. 3A , a first real estate consumer (user 1) is interested in real estate sales 1 to 4, a second real estate consumer (user 2) is interested in real estate sales 2, and a third real estate buyer Assume that (user 3) is interested in properties 2 and 3. A first real estate consumer (user 1) to a third real estate consumer (user 3) have an interest in the real estate sale 2 in common. At this time, assuming that the third real estate consumer (user 3) has a similar preference to the first real estate consumer (user 1), from the standpoint of the third real estate consumer (user 3), the unevaluated sale 1 and sale 4 For example, based on the evaluation provided by the first real estate consumer (user 1), the sale 1 and the sale 4 may be recommended to the third real estate consumer (user 3).

Referring to Figure 3b, the first real estate consumer (user 1) is interested in real estate sales 1, 3 and 4, the second real estate consumer (user 2) is interested in real estate sales 1 and 3, It is assumed that a third real estate consumer (user 3) is interested in the property 3 . A first real estate consumer (user 1) to a third real estate consumer (user 3) have an interest in the real estate for sale 3 in common. At this time, assuming that property 3 has a similarity to property 1, property 1 that has not been evaluated from the perspective of the third real estate consumer (user 3) may be recommended to the third real estate user (user 1).

4 is a flowchart illustrating a method for providing real estate information according to an embodiment of the present invention.

Referring to FIG. 4 , the method for providing real estate information includes collecting information about the real estate consumer and real estate sales information including the tendency information of the real estate consumer (S100); Based on the information about the real estate consumer and the real estate sale information, forming a relationship table between the real estate consumer and the real estate sale (S110), the relationship table is a weight value for each real estate sale calculated with the propensity of the real estate consumer and analyzing the similarity between real estate consumers using weights for each real estate sale for the real estate consumers in the relationship table (S120); And based on the similarity analysis result, it may include a step (S130) of providing a real estate sale suitable for the real estate consumer or at least one new real estate consumer in the relationship table.

In an embodiment, the information on the real estate consumer is information required for membership registration, and may include profile information of the real estate consumer such as name, gender, age, current residence, desired price range, desired area, square footage, and number of rooms. The information on the real estate consumer may further include propensity information of the real estate consumer that can be inferred by the search pattern and behavior pattern of the real estate consumer. Specifically, the propensity information of the real estate consumer may be analyzed through the search history of the real estate consumer, the use history of the related real estate information providing service, or a combination thereof. The above real estate sale information includes location, desired transaction type (sale/jeonse/monthly rent), expected sale period, square footage, management fee, housing type (apartment, villa, country house, etc.), housing direction, internal structure, completion date, and price information. Detailed information on buildings, including traffic safety (location of frequent traffic accident sites, child accident sites, etc.), transportation accessibility (KTX, subway, bus stop locations, etc.), convenience facilities (department store, market, mart, park location, etc.), It may include information related to the surrounding environment, such as public security facilities (police stations, CCTV locations, etc.), medical facilities (hospital locations, etc.), schools and academies, the residence of sex offenders, and areas where policing accidents are common.

Optionally, the detailed personal information of the real estate consumer may be utilized to form a relationship table between the real estate buyer and the real estate for sale. The detailed personal information of the real estate consumer may be defined as personal information that can be viewed and provided through an authentication means such as property, marital status, number of family members, wages, and card details. The authentication means may include public authentication or authentication through a mobile phone or e-mail that can confirm whether it is an individual.

In one embodiment, real estate consumers in the relationship table can be subdivided through the personal details, form a group with a similar tendency, and real estate buyers who form a group with a similar tendency are interested in real estate listings in the relationship table It may include real estate listings that have

In one embodiment, the similarity between the real estate consumers is a Pearson correlation coefficient (pearson correlation coefficient), cosine similarity (cosine similarity), adjusted cosine similarity (adjusted cosine similarity), mean squared difference similarity (mean squared difference similarity), jacquard similarity (jaccard similarity) or a combination thereof.

In an embodiment, a weight value that is not calculated in the relation table may be predicted through a supervised learning algorithm. The supervised learning algorithm includes K-Nearest Neighbors, Linear Regression, Logistic Regression, Support Vector Machine (SVM), Decision Tree and Random It may include any one of a forest (Random Forest) and an artificial neural network (Artificial Neural Network).

In one embodiment, the method may further include updating the weight of the real estate for sale in the relationship table by reflecting a review or rating for a real estate sale of a current resident or previous other real estate consumer of the real estate for sale. By updating the weight value based on a review or rating based on the experience of the current resident or previous real estate consumer of the real estate sale, it is possible to provide information that is difficult to understand without the real estate buyer directly living. For example, information such as whether it is safe at night because there are many street lights and surveillance cameras around, that the apartment community is active, that many young people in the apartment complex seem lively, etc. can also be considered in real estate transactions.

In one embodiment, predicting the future real estate price of the real estate sale through the learned artificial intelligence engine; And the step of providing information about the future real estate price of the predicted real estate for sale to the real estate consumer may be further included.

The artificial intelligence engine can be learned from past real estate price data, such as past market price, actual transaction price, transaction volume and corresponding price per area (eg, per pyeong), and historical real estate market price trends and growth rates that can be known through this. And the artificial intelligence engine may predict real estate price prediction data that predicts what the price of real estate will be in the future, based on the learned past real estate price data.

In addition, rather than performing learning using only the past real estate price data, the artificial intelligence engine performs learning by further using data that has influenced the price of real estate together with the past real estate price data, that is, the past real estate impact data. It can improve the accuracy of prediction.

According to an embodiment, the historical real estate impact data may mean urban development plan data for each region. In general, since urban development planning can have a large impact on price fluctuations of real estate, it can be an important factor related to historical real estate price data that the AI engine will learn. For example, the real estate price in a given area due to factors such as a new town being built in a specific area, a subway station passing in a specific area, a new road being built, or large-scale facilities such as education/medical/large shopping malls being built This may change.

According to an embodiment, the historical real estate impact data may further include various data that may affect real estate prices as well as the urban development plan data. For example, the historical real estate impact data may include historical article data determined to have had an influence on real estate prices in the past. The past article may be, for example, an article about the fact that a lot of noise is generated in a specific area, natural disasters such as earthquakes and typhoons frequently occur, or the crime rate is increasing. After such articles are published in a particular area, demand for real estate in that area may decrease, which may result in a fall in property prices. Conversely, after a positive article is published in a specific region, demand for real estate in that region increases, and real estate prices may rise.

Accordingly, the historical real estate impact data may include data on real estate-related policies, such as urban development plans, as well as data on influential facts, such as the aforementioned past articles.

As described above, using an artificial intelligence engine learned using big data, an active strategic portfolio can be developed so that real estate buyers can achieve real estate-related target conditions (eg, possession of a specific real estate, occupancy status, etc.) It has the effect of guiding the real estate buyer to reach the target state effectively.

In one embodiment, the real estate for sale may include any one of a first real estate for sale currently for sale and a second real estate for sale likely to come out in the next several months. As such, by recommending the second real estate for sale, which is likely to come out for sale within the next several months, to the real estate consumer, it is possible to provide the real estate buyer with a wider variety of options.

It is common in the art to which the present invention pertains that the present invention described above is not limited to the above-described embodiments and the accompanying drawings, and various substitutions, modifications and changes are possible within the scope without departing from the technical spirit of the present invention. It will be clear to those who have the knowledge of

Claims (18)

As a real estate information providing device,
one or more processors; and
contains a database;
The processor is
Collecting and storing information about the real estate consumers and real estate sales information, including the tendency information of real estate consumers, in the database,
Based on the information on the real estate buyers and the real estate sale information, a relationship table between the real estate buyers and the real estate sale information is formed, and the relationship table is a weight value for each real estate sale calculated by the propensity of the real estate consumers including,
Analyzing the similarity between real estate buyers using the weights for each real estate sale for the real estate consumers in the relationship table,
Determine a plurality of real estate buyers having a similar preference for real estate for sale based on the analysis result of the similarity between real estate consumers, and information on at least one real estate sale preferred by a first real estate consumer among the plurality of real estate consumers A real estate information providing device for controlling to recommend a second real estate consumer different from the first real estate consumer among the plurality of real estate consumers. 2. The method of claim 1
The similarity between the real estate consumers is a Pearson correlation coefficient, cosine similarity, adjusted cosine similarity, mean squared difference similarity, jaccard similarity or Real estate information providing device determined from a combination thereof. 2. The method of claim 1
The propensity information of the real estate consumer is determined from the personal details that can be viewed and provided based on the membership registration information of the real estate information providing service, the usage pattern of the real estate information providing service, the search pattern or authentication of related real estate information, or a combination thereof Real estate information providing device. 4. The method of claim 3
Real estate consumers in the relationship table can be subdivided through the personal details, and form groups with similar tendencies,
Real estate information providing apparatus including real estate for sale in which real estate buyers who form a group having a similar tendency are interested in real estate for sale in the relationship table. 2. The method of claim 1
Weight values not calculated in the relation table are predicted through a supervised learning algorithm,
The supervised learning algorithm includes K-Nearest Neighbors, Linear Regression, Logistic Regression, Support Vector Machine (SVM), Decision Tree and Random Real estate information providing device including any one of forest (Random Forest), artificial neural network (Artificial Neural Network). 2. The method of claim 1
the processor
Real estate information providing apparatus for updating the weight of the real estate sale in the relationship table by reflecting a review or rating of at least one real estate sale among current residents and other real estate consumers of the real estate sale. 2. The method of claim 1
the processor
Predicting the future real estate price of the real estate sale through the learned artificial intelligence engine,
Real estate information providing device for providing information about the future real estate price of the predicted real estate sale to the real estate consumer. 2. The method of claim 1
The real estate for sale is a real estate information providing device including any one of a first real estate for sale currently for sale and a second real estate for sale that is likely to come out for sale in the next several months. 2. The method of claim 1
The information on the real estate consumer includes profile information of the real estate consumer required at the time of membership registration and detailed personal information of the real estate consumer,
The propensity information of the real estate consumer is determined through the search history of the real estate consumer, the related real estate information providing service use history, or a combination thereof,
The real estate sale information includes detailed information about the building including location, desired transaction type, expected sale period, square footage, management fee, house type, housing direction, internal structure, completion date, price information, and traffic safety, traffic accessibility, convenience facilities, A real estate information providing device that includes information related to the surrounding environment, such as public security facilities, medical facilities, schools and academies, the residence of sex offenders, and areas where policing accidents are common. 10. The method of claim 9
The detailed personal information of the real estate consumer is a real estate information providing device including personal information that can be viewed and provided through an authentication means. Receiving, by the real estate information providing device, information on the real estate consumers including propensity information of real estate consumers and real estate sales information from the real estate consumers through a communication interface module;
Forming, by the real estate information providing device, a relationship table between the real estate buyers and the real estate sales information based on the information on the real estate buyers and the real estate sales information using a processor, the relationship table is the real estate including a weight value for each real estate sale calculated by the propensity of consumers;
analyzing, by the real estate information providing apparatus, a degree of similarity between real estate consumers using the processor using weights for each real estate sale for the real estate consumers in the relation table;
determining, by the real estate information providing device, a plurality of real estate consumers having similar preferences for real estate for sale based on a similarity analysis result between the real estate consumers using the processor; and
The real estate information providing device provides information on at least one real estate for sale preferred by a first real estate consumer among the plurality of real estate consumers to a second real estate consumer different from the first real estate consumer among the plurality of real estate consumers A method of providing real estate information comprising the step of making a recommendation through a communication interface module. 12. The method of claim 11
The similarity between the real estate consumers is a Pearson correlation coefficient, cosine similarity, adjusted cosine similarity, mean squared difference similarity, jaccard similarity or A method of providing real estate information determined from a combination thereof. 12. The method of claim 11
The propensity information of the real estate consumer is determined from the personal details that can be viewed and provided based on the membership registration information of the real estate information providing service, the usage pattern of the real estate information providing service, the search pattern or authentication of related real estate information, or a combination thereof How to provide real estate information. 14. The method of claim 13
Real estate consumers in the relationship table can be subdivided through the personal details, and form groups with similar tendencies,
A method of providing real estate information including real estate for sale in which real estate buyers who form a group having a similar tendency are interested in real estate for sale in the relationship table. 12. The method of claim 11
Weight values not calculated in the relation table are predicted through a supervised learning algorithm,
The supervised learning algorithm includes K-Nearest Neighbors, Linear Regression, Logistic Regression, Support Vector Machine (SVM), Decision Tree and Random A method of providing real estate information including any one of a forest (Random Forest) and an artificial neural network (Artificial Neural Network). 12. The method of claim 11
The real estate information providing method further comprising the step of updating the weight of the real estate for sale in the relationship table by reflecting the reviews or ratings for the real estate sale of the real estate consumer and the current resident of the real estate sale. 12. The method of claim 11
predicting the future real estate price of the real estate sale through a learned artificial intelligence engine; and
Real estate information providing method further comprising the step of providing information about the future real estate price of the predicted real estate sale to the real estate consumer. 12. The method of claim 11
The real estate for sale is a method of providing real estate information including any one of a first real estate for sale currently for sale and a second real estate for sale that is likely to come out for sale in the next several months.

Images