KR102656643B1 - 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 - Google Patents

Abstract

The present invention relates to a method of recommending real estate listings using an artificial intelligence-based interactive chatbot (ChatBot). The method of the present invention includes the steps of receiving a message containing information on preference conditions for real estate listings from a user terminal, identifying one or more words contained in the received message, and dividing the received preference condition information by at least one condition. Classifying and analyzing classified conditions, generating recommended real estate listing information including at least one real estate listing corresponding to or similar to the analyzed information based on one or more identified words, received messages and generated It may include storing recommended real estate listing information in a database and transmitting at least one generated recommended real estate listing information to a user terminal.

Description

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

The present invention relates to a method and device for recommending real estate listings using an artificial intelligence-based interactive chatbot (ChatBot). More specifically, it relates to an artificial intelligence-based method that analyzes real estate preference condition information received from users and recommends corresponding real estate listings. This relates to a method and device for recommending real estate listings using an interactive chatbot (ChatBot).

The year of completion and area of apartments are standardized for each complex. On the other hand, villas have various shapes such as area and structure, so standardization is difficult.

Due to this, even within the same area, due to practical difficulties in comparing jeonse and monthly rent prices and understanding the market price, fraud due to jeonse fraud and false listings are frequently occurring.

To prevent problems like these, valuation standards are needed for all housing types, and a system is needed to automatically identify false listings by standardizing data.

In addition, a method is needed to easily find real estate listings that meet the user's desired conditions based on verified information that identifies false listings.

[Patent Document 1] Korean Patent No. 10-2515560. 2023.03.24. registration.

The present invention was created to solve the above-mentioned problems, and its purpose is to provide a method and device for recommending real estate listings using an artificial intelligence-based interactive chatbot (ChatBot).

The objects of the present invention are not limited to the objects mentioned above, and other objects not mentioned can be clearly understood from the description below.

In order to achieve the above objectives, a method for recommending real estate listings using an artificial intelligence-based interactive chatbot (ChatBot) according to an embodiment of the present invention is disclosed. The method includes receiving a message containing information on preferred conditions for real estate listings from a user terminal, identifying one or more words contained in the received message, classifying the received preferred condition information by at least one condition, and , Analyzing classified conditions, Based on one or more words identified, generating recommended real estate listing information including at least one real estate listing corresponding to or similar to the analyzed information, Received message and generated real estate listing It may include storing recommendation information in a database and transmitting at least one generated real estate recommendation listing information to a user terminal.

In addition, according to one embodiment of the present invention, the step of receiving a message includes transmitting an interface for inputting choices and answers including preference conditions for real estate listings to the user terminal, and transmitting the preferred conditions for real estate listings from the user terminal. and receiving a message containing a response, wherein the preferred conditions consist of information necessary for real estate transactions, and include at least one of workplace information, neighborhood living facility information, information about the internal structure of the house, contract type information, and hate facility information. It may contain one or more pieces of information.

In addition, according to an embodiment of the present invention, the preference condition has an input box for assigning a weight to at least one condition among the preference conditions, and may include an interface for selecting by priority or selectively entering a weight proportion for each condition. there is.

In addition, according to one embodiment of the present invention, the step of generating recommended real estate listing information includes searching for at least one real estate listing information that corresponds to or is similar to the analyzed information based on one or more identified words, and searching for at least one searched real estate listing information. It may include verifying whether the real estate listing information meets one or more words of the user and received preference condition information, and generating recommended real estate listing information including at least one verified real estate listing information.

In addition, according to an embodiment of the present invention, the step of transmitting to the user terminal includes generating a message displaying the image of the generated real estate recommendation and real estate conditions, and generating a response message corresponding to the identified user intent. And it may include transmitting the generated message and response message to the user terminal.

Additionally, according to one embodiment of the present invention, the step of generating a message may be characterized by emphasizing and displaying the image of the generated real estate recommendation and the weight condition selected from the user terminal.

In addition, according to one embodiment of the present invention, the step of storing in the database may be characterized in that all histories of conversation messages and response messages transmitted and received from the user terminal are classified by real estate sale or user and stored cumulatively in the database. You can.

A real estate listing recommendation device using an artificial intelligence-based interactive chatbot (ChatBot) according to an embodiment of the present invention is disclosed. The device includes a message receiver that receives a message containing information on preferred conditions for real estate listings from a user terminal, identifies one or more words contained in the received message, and classifies the received preferred condition information by at least one condition, , a message analysis unit that analyzes classified conditions, a real estate recommendation listing information generation unit that generates real estate recommendation listing information including at least one real estate listing that corresponds to or is similar to the analyzed information, based on one or more identified words, reception, It may include a storage unit that stores the generated message and the generated real estate product recommendation information in a database, and a data transmission unit that transmits the generated at least one real estate product recommendation information to the user terminal.

In addition, according to an embodiment of the present invention, receiving a message containing information on preference conditions for real estate listings from a user terminal, identifying one or more words contained in the received message, and receiving preference condition information Classifying by at least one condition and analyzing the classified conditions; Based on one or more words identified, generating recommended real estate listing information including at least one real estate listing corresponding to or similar to the analyzed information; The method includes storing the received message and the generated real estate product recommendation information in a database and transmitting the generated at least one real estate product recommendation information to a user terminal. It can be a recording medium that can be used.

Specific details for achieving the above objectives will become clear by referring to the embodiments described in detail below along with the attached drawings.

However, the present invention is not limited to the embodiments disclosed below, and may be configured in various different forms. In order to ensure that the disclosure of the present invention is complete, those skilled in the art ( It is provided to fully inform those skilled in the art of the invention of the scope of the invention.

According to an embodiment of the present invention, the accuracy of determining false real estate listings can be improved by identifying false real estate listings using a learned artificial intelligence model.

In addition, according to an embodiment of the present invention, by filtering out false real estate listings in advance through an automatic identification method for false real estate listings using an artificial intelligence model, real estate fraud damage can be prevented in advance, and further, a healthy real estate environment can be created. You can contribute to doing this.

In addition, according to an embodiment of the present invention, information that meets the user's desired conditions can be easily and quickly searched through an artificial intelligence-based conversational chatbot (ChatBot) using information verified through identification of false items.

The effects of the present invention are not limited to the effects described above, and potential effects expected by the technical features of the present invention can be clearly understood from the description below.

The effects of the present invention are not limited to the effects described above, and potential effects expected by the technical features of the present invention can be clearly understood from the description below.

BRIEF DESCRIPTION OF THE DRAWINGS Some of the embodiments are shown in the accompanying drawings so that the above-mentioned features of the present invention may be understood in detail and a more detailed description with reference to the following embodiments. Additionally, similar reference numbers in the drawings are intended to refer to identical or similar functions across various aspects. However, note that the attached drawings only illustrate certain typical embodiments of the present invention and are not to be considered as limiting the scope of the present invention, and other embodiments having the same effect can be fully recognized. Let's do it.
Figure 1 is a schematic diagram of a device for automatically determining false real estate listings using an artificial intelligence model according to an embodiment of the present invention.
Figure 2 is a block diagram of a device for automatically determining false real estate listings using an artificial intelligence model according to an embodiment of the present invention.
Figure 3 is a flowchart of a method for automatically determining false real estate listings using an artificial intelligence model according to an embodiment of the present invention.
Figure 4 is an example table in which scores are calculated according to section settings for each element through the false listing determination step according to an embodiment of the present invention.
Figure 5 is an example of the final result through an automatic real estate false listing detection device using an artificial intelligence model according to an embodiment of the present invention.
Figure 6 is a block diagram of a real estate listing recommendation device using an artificial intelligence-based interactive chatbot (ChatBot) according to an embodiment of the present invention.
Figure 7 is a flowchart of a method for recommending real estate listings using an artificial intelligence-based interactive chatbot (ChatBot) according to an embodiment of the present invention.
Figure 8 is an example diagram of a method for recommending real estate listings using an artificial intelligence-based interactive chatbot (ChatBot) according to an embodiment of the present invention.
Figure 9 is an example of a real estate safety checklist using an artificial intelligence-based conversational chatbot (ChatBot) according to an embodiment of the present invention.

Since the present invention can be subject to various changes and can have various embodiments, specific embodiments will be illustrated in the drawings and described in detail.

The various features of the invention disclosed in the claims may be better understood by consideration of the drawings and detailed description. The apparatus, method, manufacturing method, and various embodiments disclosed in the specification are provided for illustrative purposes. The disclosed structural and functional features are intended to enable those skilled in the art to specifically implement various embodiments, and are not intended to limit the scope of the invention. The disclosed terms and sentences are intended to easily explain various features of the disclosed invention and are not intended to limit the scope of the invention.

In describing the present invention, if it is determined that a detailed description of related known technology may unnecessarily obscure the gist of the present invention, the detailed description will be omitted.

Hereinafter, a device and method for automatically determining false real estate listings using an artificial intelligence model according to an embodiment of the present invention and a computer-readable recording medium recording a program for executing the same will be described and an artificial intelligence-based interactive chatbot (ChatBot) will be described. Describes the real estate listing recommendation method, device, and computer-readable recording medium that records the program for executing the method.

The real estate safety brokerage platform system verifies information about real estate listings received from outside to determine whether they are false listings, and based on the verified real estate listing information, uses an artificial intelligence interactive chatbot to select real estate that meets the user's desired conditions. It is a system that recommends properties for sale.

Figure 1 is a schematic diagram of a device for automatically determining real estate false listings using an artificial intelligence model according to an embodiment of the present invention, and Figure 2 is a block diagram of a device for automatically determining false real estate listings using an artificial intelligence model according to an embodiment of the present invention. It's a degree.

Referring to Figures 1 and 2, the real estate false listing automatic determination device 100 using an artificial intelligence model may include a real estate data collection unit 110, a real estate false listing determination unit 130, and a false listing display unit 150. You can.

In one embodiment, the device 100 for automatically determining real estate false listings using an artificial intelligence model receives transaction type and desired transaction price information from an external server (e.g., DB sharing service dedicated to real estate agents, community or various portal real estate brokerage services, etc.). We collect information including real estate listing information, real estate listing information, and real estate transaction information, and input the collected real estate listing information into an artificial intelligence model that judges whether the registered real estate is a false listing. It is a device that reflects the information judged to be a false listing in the relevant real estate listing information and allows the application user to check the contents of the real estate that is determined to be a false listing.

In one embodiment, the real estate data collection unit 110 may collect information including real estate listing information and actual real estate transaction information including transaction type and desired transaction price information from an external server. For example, when real estate listing information is registered from the server of a DB sharing service exclusively for licensed real estate agents, a community or portal real estate brokerage service, or when real estate transaction information is registered from the Ministry of Land, Infrastructure and Transport's actual transaction price public server, the real estate data collection unit of the device ( 110) can collect the above information.

Here, real estate listing information includes transaction type information such as sale, lease, or monthly rent, and may include information affecting real estate transactions. Specifically, information that affects real estate transactions can be divided into internal factors of the real estate itself and external factors about the surrounding environment of the real estate. Internal factors include the year of completion of the real estate, walking distance to the station, exclusive use area, etc. , external factors may include information such as transaction type, price, area adjacent to the subway station, surrounding infrastructure (e.g., hospital, police station, park, entertainment district, school, supermarket, etc.) facilities, and information on future redevelopment potential.

In one embodiment, the real estate false listing determination unit 130 inputs the real estate listing information collected in the real estate data collection unit 110 into an artificial intelligence model previously learned by the user and determines the false listing using the result value. can do.

In one embodiment, if the real estate listing information registered through the real estate false listing determination unit 130 is determined to be a false listing, the false listing display unit 150 may display information reflecting the fact that the real estate listing information is a false listing. You can.

Figure 3 is a flow chart of a method for automatically determining real estate false listings using an artificial intelligence model according to an embodiment of the present invention, and Figure 4 is a score according to section setting for each element through the false listing determination step according to an embodiment of the present invention. This is an example of a table calculated, and Figure 5 is an example of the final result through an automatic real estate false listing detection device using an artificial intelligence model according to an embodiment of the present invention.

Referring to Figures 3 to 5, the method for automatically determining real estate false listings (S200) using an artificial intelligence model includes a real estate data collection step (S210), a real estate false listing determination step (S230), and a false listing display step (S250). can do.

In one embodiment, the real estate data collection step (S210) may be a step of collecting information including real estate listing information and real estate transaction information including transaction type and desired transaction price information from an external server.

In one embodiment, the real estate false listing determination step (S230) determines the false listing using the result obtained by inputting the real estate listing information collected in the real estate data collection step (S210) into an artificial intelligence model previously trained by the user. This may be a step.

More specifically, in the real estate false listing determination step (S230), the first false listing judgment is made by extracting and classifying real estate listing information by at least one element and determining the degree of real estate false listing through scoring for each classified element. After the step and the first false listing determination step, a second false listing determination step may be included in which the value of the determined degree of real estate false listing is added in consideration of the time information elapsed from the registration date of the real estate listing information.

In addition, the real estate false listing determination step (S230) may be characterized by assigning different weights depending on the transaction type of the real estate transaction or the period of time elapsed from the date of registration of real estate listing information.

In other words, the real estate false listing judgment step (S230) extracts false listing judgment elements from real estate listing information, makes a primary judgment as to whether it is a false listing according to preset standards, and then adds the time elapsed factor since the real estate listing was registered. This is the stage where the first judgment of false listing is adjusted by reflecting it.

For example, referring to Figure 4, elements of completion year, walking distance to the station, exclusive area, transaction type, price, and region are extracted from the collected real estate listing information, and each element is assigned a score by setting a certain section, It is possible to determine whether a product is a false listing based on the total score given to each element. In addition, the calculated score can be subdivided by region and nationwide to set a standard value for comparison. The above example is only an example for explaining the present disclosure, and the present disclosure is not limited thereto.

In addition, the first false listing determination step is an element-specific information extraction step of extracting at least one element-specific information that affects real estate transactions from the collected real estate listing information, and setting a range score for the extracted element-specific information in certain sections. It may include a step of calculating the score of the collected real estate listing information based on the scores for each set element, and a determination of the degree of false listing by assigning a probability corresponding to the calculated score to determine the degree of the false listing. .

In addition, the element-specific information extraction step may be a step of extracting at least one element of transaction type information, price information, address information to determine the distance to public transportation facilities, area, and year of completion from the collected real estate listing information. there is.

Additionally, the step of setting a score may be a step of setting a score by dividing the information for each element extracted through the information for each element into certain sections. For example, in the case of completion year, points can be given in three categories: 4 years or less, 5 to 10 years or less, and 11 years or more, and the walking distance to the station is 5 minutes or less, 6 to 10 minutes or less, and 11 minutes or more. Points can be given by dividing into 3 sections. In the case of exclusive area, points can be given by dividing into 3 sections: 5 pyeong or less, 6 to 10 pyeong or less, and 11 pyeong or more. In the case of transaction type, it can be divided into sale, jeonse, and monthly rent. Price can be divided into three categories: less than 200 million won, 200 million to 300 million won, and more than 300 million won. In the case of region (subway station), points can be given based on the area where the subway station is located. The score set by determining a certain range for each element can be weighted additionally to the score in the range depending on the transaction type. The above example is only an example for explaining the present disclosure, and the present disclosure is not limited thereto.

Additionally, the step of calculating the score may be a step of calculating the score of the collected real estate listing information based on the score for each element set in the score setting step.

Additionally, the step of determining the degree of false listing may be a step of determining the degree of false listing by assigning a probability corresponding to the score calculated in the score calculation step.

In addition, the stage of determining the degree of false listing may be characterized in that the higher the score calculated in the score calculation stage, the higher the degree of false listing. In other words, in the stage of determining the degree of false listing, the higher the score of the real estate listing, the more likely it is that the real estate that everyone wants to sign a contract with. However, considering the transaction reality, the real estate that everyone likes should be contracted in a short period of time, but if not, there is a high probability that it is a false listing. I think it's high.

In addition, the step of determining the degree of false listing may be a step of determining the degree of false listing of the real estate listing information based on the results of analyzing actual real estate transaction information similar to the real estate listing information. In other words, by analyzing the actual transaction information of similar real estate and using it as a comparison when determining whether a real estate registration is a false listing, it reflects that real estate in the same area, same complex, similar time, and same area is highly likely to be traded under similar conditions. You can.

In addition, by analyzing actual transaction prices and comparing them with real estate listings, overheating of real estate transactions can be prevented, and through this, a sound real estate transaction order can be maintained.

In addition, in the stage of determining the level of false listings, the real estate listing information that is likely to be traded in the future can be estimated by analyzing the collected real estate transaction information and monitoring changes in the actual transaction information, and the registered real estate listing information that is most similar to the estimated real estate listing information can be estimated. This may be a step to identify properties with high trading potential by comparing and analyzing property information. In addition, even though it is a property judged to have a high possibility of being traded, if time passes from the date of registration, it can be determined that there is a high probability that it is a false property.

Additionally, the second false listing determination step may be a step of adding a value to the already determined degree of real estate false listing by considering the time information elapsed from the registration date of the real estate listing information after the first false listing determination step. In other words, the purpose is to re-determine the false listing information determined at the time of registration by considering the time information that has elapsed from the registration date. Specifically, the second false listing determination step is to determine the degree of false listing by first judging real estate false listings through listing information. After the judgment has been made, the transaction may continue to be reflected in a state where the transaction has not been completed even though time has passed since the listing reflecting the relevant fact was registered.

More specifically, the second false listing determination step is the real estate judged by considering the number of visits to the real estate listing and the number of visit rejection notifications for visit requests compared to the time elapsed from the registration date of the real estate listing information after the first false listing determination step. This may be the step of adding up the number of false listings. In other words, the purpose is to increase the accuracy of judging false listings by reflecting information such as whether there are a lot of visitors to a real estate listing and when a visit reservation is refused.

In addition, the second false listing determination step may further include a step of readjusting the numerical value by reflecting visitor feedback from real estate listing visitors. For example, in the second false listing determination step, additional judgment can be made through feedback from false listing reports and visit reviews from customers who actually visited the relevant listing. In addition, in the second false listing determination step, an additional judgment can be made by checking with the real estate agent scheduled to visit whether the property to be visited can actually be visited on the same day that the house is scheduled to be visited. The above example is only an example for explaining the present disclosure, and the present disclosure is not limited thereto.

In one embodiment, the false listing display step (S250) displays information reflecting the content determined to be a false listing in the real estate listing information when the real estate listing information determined through the real estate listing false listing determination step (S230) is determined to be a false listing. This may be a step. For example, referring to Figure 6, the display stage can be expressed by visualizing the transaction type, price standard, station walking distance, area, and subway station elements by setting the range set in each of the transaction type, price standard, station walking distance, area, and year of completion. This may be a step in which the results of the judgment are visualized in a graph so that the user can easily determine whether the product is a false product.

The method for automatically determining false real estate listings using an artificial intelligence model according to an embodiment of the present invention can also be implemented in the form of a computer-readable recording medium that records a program for execution on a computer. Computer-readable recording media can be any available media that can be accessed by a computer, and can include both volatile and non-volatile media, removable and non-removable media. Additionally, computer-readable recording media may include computer storage media. Computer storage media may include both volatile and non-volatile, removable and non-removable media implemented in any method or technology for storage of information such as computer-readable instructions, data structures, program modules, or other data.

Exemplary modules, steps, or combinations thereof of the embodiments described herein may be implemented by electronic hardware (digital designs designed by coding, etc.), software (various types of applications including program instructions), or combinations thereof. It can be implemented. Whether it is implemented in hardware and/or software may vary depending on design constraints imposed on the user terminal.

Figure 6 is a block diagram of a real estate listing recommendation device using an artificial intelligence-based interactive chatbot (ChatBot) according to an embodiment of the present invention.

Referring to Figure 6, the real estate listing recommendation device 1000 using an artificial intelligence-based interactive chatbot (ChatBot) includes a message receiving unit 1100, a message analysis unit 1300, a real estate recommendation listing information generating unit 1500, and a storage unit. It may include (1700) and a data transmission unit (1900).

In one embodiment, the message receiver 1100 may receive a message containing information on preferred conditions for real estate listings from a user terminal. When the user executes the program through the application, a selection interface for preset real estate preference conditions is transmitted to the user terminal, and the user terminal receiving the transmission can select the preferred conditions and transmit them to the device.

In one embodiment, the message analysis unit 1300 identifies one or more words contained in a message received through the message receiver 1100, classifies the received preference condition information into at least one condition, and classifies the classified conditions. It can be analyzed.

More specifically, the message analysis unit 1300 can identify one or more words contained in the message by analyzing it using an artificial intelligence model. Through this, the device can identify one or more words included in the message, classify condition information by condition, and analyze the classified conditions.

In one embodiment, the real estate recommendation listing information generation unit 1500 generates real estate recommendation listing information including at least one real estate listing that corresponds to or is similar to the analyzed information (analyzed conditions) based on one or more identified words. can do.

That is, the real estate recommendation listing information generation unit 1500 searches for real estate listings that correspond to or are similar to the conditions, and based on one or more words identified in the message analysis unit 1300, the searched information is combined with one or more words and preferences of the user. After verifying that the conditions are met, recommended real estate listing information can be generated to be transmitted to the user terminal.

In one embodiment, the storage unit 1700 may store the received message and the generated real estate listing recommendation information in a database. More specifically, the storage unit 1700 can store all conversation contents transmitted and received with the user terminal in time series, and can also store all conversation contents with other user terminals about the same product. By storing not only the content of real estate listings but also all conversations in a time series, the user's intentions can be understood more clearly and preparation can be made in case a contract dispute arises in the future.

In one embodiment, the data transmission unit 1900 may include a data transmission unit that transmits at least one real estate recommendation listing information generated by the real estate recommendation listing information generating unit 1500 to the user terminal.

In addition, the real estate listing recommendation device 1000 using an artificial intelligence-based interactive chatbot (ChatBot) allows the chatbot to communicate between the user, real estate agent, and landlord after recommending a real estate listing or regardless of whether or not the real estate listing is recommended. By analyzing the conversation history, it can be determined whether the items on the safety checklist have been satisfied, and for items not confirmed on the checklist, a notification can be sent to the user.

Additionally, the message receiver 1100 may include a safety checklist containing items to be careful about when signing a real estate contract from the user terminal when receiving preference condition information for real estate listings. By checking the precautions in advance when signing a contract through the safety checklist, users can make a more reliable and safe transaction when visiting real estate or making a final contract.

For example, the safety checklist includes matters to pay attention to when signing a real estate contract, such as non-payment of landlord tax, confirmation of the landlord matching the copy of the register, registration information at the real estate agent, confirmation of normal business, confirmation of no changes in registration before signing the contract, and confirmation of whether there are any changes to the registration before signing the contract. It may include information such as whether the city housing rental standard contract is used. The above example is only an example for explaining the present disclosure, and the present disclosure is not limited thereto.

In addition, the real estate listing recommendation device 1000 using an artificial intelligence-based interactive chatbot (ChatBot) recommends real estate listings and then the chatbot analyzes the conversation history between the user, the real estate agent and the landlord, and creates a chatbot ( As a result of ChatBot's analysis of electronic contracts or contract scans and conversation history between users and real estate agents and landlords, if the contract contents and special terms and conditions in the contract are abnormal or the process of carrying out the special terms and conditions is abnormal, the user is at risk. Notifications can be sent.

For example, in cases where special contract conditions are abnormal, analysis of conversation content shows that the costs borne by the tenant for wallpapering, cleaning, etc. far exceed the average cost borne relative to the area (exceeding 30% of the average), or the costs required for management fees are 30% higher than the average cost compared to the area. Cases where the cost far exceeds the average burden can be judged as abnormal. The above example is only an example for explaining the present disclosure, and the present disclosure is not limited thereto.

For example, if the process of carrying out the special contract conditions is abnormal, it can be found in the conversation content. If the special contract is “If the house is mortgaged by a bank, transfer after repaying the loan before registration of transfer of ownership,” the loan amount can be deposited in the conversation history. It can be judged as an abnormal case if there is a request to deposit money to the landlord or a third party rather than directly to the lending bank.

In addition, although not shown, the real estate listing recommendation device 1000 using an artificial intelligence-based interactive chatbot (ChatBot) recommends real estate listings or after real estate brokerage ends, regardless of whether real estate listings are recommended ( (1) As a necessary part immediately after brokerage (or after the transaction under a real estate contract is completed), moving-related information (e.g., moving center, moving costs, etc.) and cleaning information (e.g., cleaning company, cleaning cost, etc.) appropriate to the relevant property information ), etc., using an artificial intelligence-based interactive chatbot, or (2) property repair information (e.g. water/sewer repair information, electrical repair information, etc.), information on necessary services related to the property, etc., in preparation for actual use after brokerage. can be recommended using an artificial intelligence-based conversational chatbot. At this time, the chatbot analyzes the conversation history of other users and provides information on properties contracted by other users (e.g. moving date, address of residence before moving, area of residence before moving, address of contract property, area of contract property, number of contract properties) /Sewage breakdown, electrical failure of the contracted property, etc.), moving information selected after the contract, cleaning information, property repair information, necessary service information related to the property, etc. can be learned, and based on the learned information, the contracted property can be selected. Once relevant data is entered, moving-related information, cleaning information, property repair information, property-related necessary service information, etc. can be output to the user. The above example is only an example for explaining the present disclosure, and the present disclosure is not limited thereto.

Figure 7 is a flow chart of a method for recommending real estate listings using an artificial intelligence-based interactive chatbot (ChatBot) according to an embodiment of the present invention, and Figure 8 is an artificial intelligence-based interactive chatbot (ChatBot) according to an embodiment of the present invention. It is an example diagram of a method of recommending real estate listings using , and Figure 9 is an example of a real estate safety checklist using an artificial intelligence-based interactive chatbot (ChatBot) according to an embodiment of the present invention.

7 to 9, the real estate listing recommendation method (S2000) using an artificial intelligence-based interactive chatbot (ChatBot) includes receiving a message (S2100), identifying one or more words (S2200), and setting conditions. It may include an analysis step (S2300), a step of generating recommended real estate listing information (S2400), a step of storing (S2500), and a step of transmitting to the user terminal (S2600).

In one embodiment, the step of receiving a message (S2100) may include receiving a message containing information on preferred conditions for real estate listings from a user terminal.

More specifically, the step of receiving a message (S2100) includes transmitting an interface for inputting choices and answers including preference conditions for real estate listings to the user terminal, and including preference conditions and answers for real estate listings from the user terminal. This may be the stage of receiving a message.

Here, the preferred conditions for real estate sales consist of information necessary for real estate transactions, including workplace information and neighborhood living facility information (e.g., transportation, hospitals, supermarkets, parks, sports facilities, daycare centers, kindergartens, schools, police stations, fire stations, etc.) , it can include at least one of information about the internal structure of the house, contract type information, and hate facility information, and other additional conditions requested by the user can be entered.

Additionally, the preferred conditions for real estate listings may assign weight to at least one condition among the preferred conditions, and may include an interface for prioritizing conditions or selectively inputting weight proportions for each condition.

In one embodiment, the step of identifying one or more words (S2200) may be a step of identifying one or more words of the user included in the message received in the step of receiving the message (S2100). In the step of identifying one or more words (S2200), one or more words contained in the message can be identified by analyzing them using an artificial intelligence model. That is, at least one word can be identified in the user's message.

In one embodiment, the step of analyzing the conditions (S2300) may be a step of classifying the received preference condition information into at least one condition and analyzing the classified conditions.

In one embodiment, the step of generating recommended real estate listing information (S2400) generates recommended real estate listing information including at least one real estate listing that corresponds to or is similar to the analyzed information (or conditions) based on one or more identified words. It may be a creation stage.

More specifically, the step of generating recommended real estate listing information (S2400) is based on the one or more words identified through the step of identifying one or more words (S2200), at least one word corresponding to or similar to the analyzed information (or condition). Searching for real estate listing information, verifying whether the searched at least one real estate listing information meets one or more words of the user and received preference condition information, and real estate recommendation listing information including the verified at least one real estate listing information. It may include the step of generating.

In one embodiment, the step of generating real estate recommendation listing information (S2400) is real estate recommendation listing information corresponding to preference condition information (i.e. input data) received using a convolutional neural network (CNN)-based artificial intelligence model (S2400). That is, it may include a step of generating output data).

More specifically, the CNN-based artificial intelligence model analyzes (1) the conversation history through the interactive chatbot, and determines the relationship between the preferred conditions selected by users among the conversation history of previous users and the properties for which visits and/or contracts have been confirmed. (2) In addition to the conversation history, the real estate preferences of a specific user are learned by learning the correlation between the preference conditions selected by users through touch input within the application and the properties for which visits and/or contracts have been confirmed. When condition information is input, it can be set to output specific property information corresponding to the preferred condition information.

In one embodiment, the step of storing in a database (S2500) may be a step of storing the received message and the generated real estate listing recommendation information in the database.

More specifically, the step of storing in the database (S2500) may be characterized by dividing all history of conversation messages and response messages transmitted and received from the user terminal by real estate sale or user and cumulatively storing them in the database.

In one embodiment, the step of transmitting to the user terminal (S2600) may be a step of transmitting at least one generated real estate recommendation listing information to the user terminal.

More specifically, the step of transmitting to the user terminal (S2600) includes generating a message displaying the image of the generated real estate recommendation and real estate conditions, generating a response message corresponding to the identified user intention, and the generated message. and transmitting a response message to the user terminal.

Additionally, in the step of generating a message (S2600), the image of the generated real estate recommendation and the weight condition selected from the user terminal can be displayed with emphasis. Additionally, in the step of generating a message (S2600), the user's preferred conditions may be displayed on the image of the recommended product.

In addition, the real estate listing recommendation method (S2000) using an artificial intelligence-based interactive chatbot (ChatBot) is a method in which the chatbot (ChatBot) communicates between the user, real estate agent, and landlord after recommending a real estate listing or regardless of whether or not the real estate listing is recommended. It may further include a step of determining whether the items in the safety checklist have been satisfied by analyzing the conversation history, and as a result of the determination, a notification may be sent to the user for items not confirmed in the checklist.

Additionally, the preferred condition information for real estate listings received in the message receiving step (S2100) may include a safety checklist in which matters to pay attention to when entering into a real estate contract are checked from the user terminal. By checking the precautions in advance when signing a contract through the safety checklist, users can make a more reliable and safe transaction when visiting real estate or making a final contract.

For example, the safety checklist includes matters to pay attention to when signing a real estate contract, such as non-payment of landlord tax, confirmation of the landlord matching the copy of the register, registration information at the real estate agent, confirmation of normal business, confirmation of no changes in registration before signing the contract, and confirmation of whether there are any changes to the registration before signing the contract. It may include information such as whether the city housing rental standard contract is used. The above example is only an example for explaining the present disclosure, and the present disclosure is not limited thereto.

In addition, the real estate listing recommendation method (S2000) using an artificial intelligence-based conversational chatbot (ChatBot) adds a step where the chatbot (ChatBot) analyzes the conversation history between the user, real estate agent, and landlord after recommending the real estate listing. In the step of analyzing the conversation history, the ChatBot analyzes the electronic contract or scanned copy of the contract and the conversation history between the user, the real estate agent, and the landlord, and as a result, the contract contents and special terms and conditions in the contract are abnormal or are not included in the special contract. If the process of carrying out the condition is determined to be abnormal, this may be a step in which a notification of a dangerous situation is sent to the user.

For example, in cases where special contract conditions are abnormal, analysis of conversation content shows that the costs borne by the tenant for wallpapering, cleaning, etc. far exceed the average cost borne relative to the area (exceeding 30% of the average), or the costs required for management fees are 30% higher than the average cost compared to the area. Cases where the cost far exceeds the average burden can be judged as abnormal. The above example is only an example for explaining the present disclosure, and the present disclosure is not limited thereto.

For example, if the process of carrying out the special contract conditions is abnormal, it can be found in the conversation content. If the special contract is “If the house is mortgaged by a bank, transfer after repaying the loan before registration of transfer of ownership,” the loan amount can be deposited in the conversation history. It can be judged as an abnormal case if there is a request to deposit money to the landlord or a third party rather than directly to the lending bank. The above example is only an example for explaining the present disclosure, and the present disclosure is not limited thereto.

The method of recommending real estate listings using an artificial intelligence-based interactive chatbot (ChatBot) according to an embodiment of the present invention can also be implemented in the form of a computer-readable recording medium recording a program for execution on a computer. Computer-readable recording media can be any available media that can be accessed by a computer, and can include both volatile and non-volatile media, removable and non-removable media. Additionally, computer-readable recording media may include computer storage media. Computer storage media may include both volatile and non-volatile, removable and non-removable media implemented in any method or technology for storage of information such as computer-readable instructions, data structures, program modules, or other data.

Exemplary modules, steps, or combinations thereof of the embodiments described herein may be implemented by electronic hardware (digital designs designed by coding, etc.), software (various types of applications including program instructions), or combinations thereof. It can be implemented. Whether it is implemented in hardware and/or software may vary depending on design constraints imposed on the user terminal.

The above description is merely an illustrative explanation of the technical idea of the present invention, and those skilled in the art will be able to make various changes and modifications without departing from the essential characteristics of the present invention.

Accordingly, the embodiments disclosed in this specification are not intended to limit the technical idea of the present invention, but are for illustrative purposes, and the scope of the present invention is not limited by these embodiments.

The scope of protection of the present invention should be interpreted in accordance with the claims, and all technical ideas within the equivalent scope should be understood to be included in the scope of rights of the present invention.

Claims (9)

In the method of recommending real estate listings using an artificial intelligence-based interactive chatbot (ChatBot) performed by an electronic device,
Receiving a message containing preference condition information for real estate listings from a user terminal;
identifying one or more words contained in the received message;
Classifying the received preference condition information into at least one condition and analyzing the classified condition;
Based on the one or more identified words, generating recommended real estate listing information including at least one real estate listing corresponding to the analyzed conditions;
Storing the received message and the recommended real estate listing information in a database;
Transmitting the real estate recommendation listing information to the user terminal;
After transmitting the real estate recommendation listing information, the conversation history of the user terminal regarding the real estate recommendation listing information is analyzed to determine whether items in a predetermined safety checklist for the real estate recommendation listing information are checked, and the predetermined Transmitting a notification about unchecked items among the items on the safety checklist to the user terminal;
The recommended real estate information is transmitted, and after the real estate brokerage for the recommended real estate information is terminated, the artificial intelligence-based interactive chatbot is used to provide moving-related information, cleaning information, and property repair information corresponding to the contracted property information. Recommendation step; and
As a result of analyzing the scanned copy of the contract for the above contract property information, the special terms and conditions in the contract are abnormal, and the cost borne by the lessee per area for cleaning costs exceeds the first threshold, and the lessee is borne per area for management costs. If the cost exceeds the second threshold, transmitting a notification about a dangerous situation to the user terminal,
The moving-related information includes a moving center and moving costs, the cleaning information includes a cleaning company and cleaning costs, the property repair information includes water/sewerage repair information and electrical repair information,
The step of receiving the message is,
Transmitting an interface for inputting choices and answers including preferred conditions for real estate listings to the user terminal; and
Comprising the step of receiving a message containing information on preferred conditions for real estate listings and a response from the user terminal,
The above preference information is,
Workplace information, neighborhood living facility information, information about the internal structure of the house, contract type information and hate facility information, information checked from the predetermined safety checklist above regarding matters to be aware of when signing a real estate contract, and weight given to the preference condition information above. There is an input box for assigning , and includes an interface for selectively inputting selection by priority or weighting by condition,
The predetermined safety checklist includes whether the landlord has not paid taxes, whether the landlord matches the registered copy, whether the real estate agency is registered, whether it is operating normally, whether there are any changes in the registration before signing the contract, and whether the standard housing rental contract is used when signing the contract,
The step of generating the real estate recommendation listing information is,
Based on the identified one or more words, searching for at least one real estate listing information corresponding to the analyzed conditions;
Verifying whether the searched at least one real estate listing information meets the one or more words and the received preference condition information; and
Generating real estate recommendation listing information including the verified at least one real estate listing information,
The step of transmitting to the user terminal is,
generating a message displaying images and real estate conditions of the generated real estate recommendation listing information and emphasizing weight conditions selected from the user terminal;
Among the generated messages, storing the entire history of conversation messages and response messages transmitted and received from the user terminal in a database to distinguish them in correspondence with the real estate recommendation information and the preference condition information;
generating a response message corresponding to the one or more identified words; and
Including transmitting the generated message and response message to the user terminal,
Real estate product recommendation method using artificial intelligence-based interactive chatbot (ChatBot).
delete delete delete delete delete According to paragraph 1,
The step of storing in the database is,
Characterized in that all history of conversation messages and response messages transmitted and received from the user terminal are classified by real estate sale or user and stored cumulatively in a database.
Real estate product recommendation method using artificial intelligence-based interactive chatbot (ChatBot).
delete delete