WO2022102965A1

WO2022102965A1 - Method for analyzing document

Info

Publication number: WO2022102965A1
Application number: PCT/KR2021/013488
Authority: WO
Inventors: 진보람; 김성열; 황지현; 문보석
Original assignee: 삼성전자주식회사
Priority date: 2020-11-10
Filing date: 2021-10-01
Publication date: 2022-05-19
Also published as: KR20220063459A

Abstract

A method for analyzing a document is disclosed. The present method for analyzing a document comprises steps of: receiving a document; extracting texts included in the document; identifying texts corresponding to a plurality of identification items from among the extracted texts; recognizing, as information corresponding to the identification items, a text written in an area adjacent to the identified texts; classifying the information corresponding to the identification items into a plurality of preset groups; processing pieces of information belonging to the plurality of groups according to processing processes set to be different by group; and providing analysis data about the document on the basis of the processing result.

Description

How to analyze documents

The present disclosure relates to a document analysis method, and more particularly, to a document analysis method for providing analysis data for a document by analyzing information meaning a layout and text included in a document.

Various documents are used in daily life, and in particular, in the case of a legal document such as a contract, the contracting parties need to determine whether the contract included in the document is secure.

For example, in the case of a real estate contract document, not only the contract details included in the contract, but also the information disclosed in the study documents such as a registered copy of the real estate for sale subject to the contract should be considered comprehensively.

However, there was a problem in that it was difficult for the general user to clearly analyze the contents of the contract document including legal information, etc., and to judge the safety of the real estate contract by considering the related information comprehensively.

The present disclosure is to solve the above problems, and an object of the present disclosure is to recognize information from a layout and text included in a document, and synthesize the information recognized from the document and information received from the outside to analyze data for the document. It is intended to provide a document analysis method provided.

In order to achieve the above object, a document analysis method according to an embodiment of the present disclosure includes the steps of receiving a document, extracting text included in the document, and corresponding to a plurality of identification items among the extracted texts Recognizing a text to be identified, recognizing a text written in an area adjacent to the identified text as information corresponding to the identification item, classifying the information corresponding to the identification item into a plurality of preset groups, for each group and processing information belonging to the plurality of groups according to a differently set processing process, and providing analysis data for the document based on the processing result.

In this case, the processing of the information belonging to the plurality of groups includes: receiving related information from an external server based on information belonging to at least one of the plurality of groups; and the related information received from the external server and generating analysis data for the document based on information recognized from the document.

In this case, the generating of the analysis data may include comparing the related information received from the external server with corresponding information among information recognized from the document to determine whether they match.

On the other hand, the document is a real estate contract document, and the generating of the analysis data includes related information received from the external server and information recognized from the document according to the property safety determination process for determining property safety through a preset algorithm can be processed to generate the analysis data.

On the other hand, the document is a real estate contract document, and the generating of the analysis data includes the related information received from the external server and the The analysis data may be generated by processing information recognized from the document.

On the other hand, the document is a real estate contract document, and the processing process includes spelling correction, spacing correction, number conversion, term conversion, related content addition, and property safety determination process, applied to the first group among the plurality of groups It includes a first processing process set to be so, spacing correction, number conversion, term conversion, accredited information match determination process, property safety determination process, and property suitability determination process based on user preference information, in the second group of the plurality of groups It includes a second processing process set to be applied to a second processing process, spacing correction, term conversion, a process for determining whether public information matches, and a third processing process set to be applied to a third group among the plurality of groups, The plurality of groups may be classified based on information required for each processing process.

In this case, the first group may include information on real estate contract details, the second group may include information indicating characteristics of real estate, and the third group may include information for specifying real estate.

On the other hand, the document is a real estate contract document, and the analysis data for the document is at least one of the accuracy of information related to the real estate and the owner specified in the real estate contract document, information on setting rights related to the real estate, and the safety of the real estate contract may include information about

In this case, the information on the safety of the real estate contract may be provided in the form of at least one of text, number, image, or video.

On the other hand, the document is a real estate contract document, and the step of receiving the related information from the external server includes a certified copy of the register of real estate specified by the document, land ledger, building ledger, cadastral map, land use plan confirmation, real estate price information, Receiving at least one piece of real estate term information from at least one external server, and generating the analysis data, generate analysis data based on the data obtained by statistical processing on the information received from the external server. can

Meanwhile, the plurality of identification items may include at least one of an address, a floor number, a lessor, an owner, a land area, a structure, a use, a building area, a deposit, a purchase price, and a special contract.

On the other hand, the step of receiving the related information from the external server, receiving the related information corresponding to each of a plurality of different time points, and the step of providing the analysis data, based on the received related information, the plurality of It is possible to generate and provide a plurality of analysis data corresponding to each time point of .

In this case, the document is a real estate contract document, and the plurality of time points may include at least two of before a real estate contract date, a real estate contract date, and after a real estate contract date.

* On the other hand, the step of extracting the text includes, if the document includes a table layout, extracting a vertical line and a horizontal line from the table layout, respectively, obtaining an intersection point between the vertical line and the horizontal line, and based on the intersection point to identify a cell included in the table layout, and an object located inside the identified cell may be extracted as text.

In this case, in the extracting of the text, the text may be extracted by applying the OCR model to the object located inside the identified cell.

On the other hand, the server device according to an embodiment of the present disclosure extracts a communication unit, a memory for storing a document, and text included in the document, and identifies texts corresponding to a plurality of identification items from among the extracted texts, a processor for recognizing a text written in an area adjacent to the identified text as information corresponding to the identification item, wherein the processor classifies the information corresponding to the identification item into a plurality of preset groups, and is different for each group The information belonging to the plurality of groups is processed according to a set processing process, and analysis data for the document is provided based on the processing result.

In this case, the processor receives related information from an external server based on information belonging to at least one of the plurality of groups, and adds the related information received from the external server and information recognized from the document. Based on the analysis data for the document may be generated.

In this case, the processor may compare the related information received from the external server with corresponding information among information recognized from the document to determine whether they match.

On the other hand, the document is a real estate contract document, and the processor processes the related information received from the external server and the information recognized from the document according to the property safety determination process for determining property safety through a preset algorithm to analyze the data can be created.

Meanwhile, in the computer-readable recording medium storing a program for analyzing a document according to an embodiment of the present disclosure, the program includes the steps of receiving a document, extracting text included in the document, and the extracted text Recognizing a text corresponding to a plurality of identification items from among, recognizing a text written in an area adjacent to the identified text as information corresponding to the identification item, and converting information corresponding to the identification item into a plurality of preset Classifying into groups, processing information belonging to the plurality of groups according to a processing process set differently for each group, and providing analysis data for the document based on the processing result.

1 is a diagram for explaining an operation of a server device that provides a document analysis method according to an embodiment of the present disclosure.

2 is a block diagram illustrating a configuration of a server device according to an embodiment of the present disclosure.

3 is a block diagram illustrating a configuration of a user terminal device according to an embodiment of the present disclosure.

4 is a diagram illustrating an example of a document to be analyzed in a document analysis method according to an embodiment of the present disclosure.

5 is a diagram for explaining an operation of extracting a vertical line from a table layout included in a document.

6 is a diagram for explaining an operation of extracting a horizontal line from a table layout included in a document.

7 is a diagram for explaining an operation of identifying cells included in a table layout by obtaining an intersection between vertical and horizontal lines extracted from the table layout.

8 is a diagram illustrating data acquired for a document.

9 is a diagram for explaining an operation of comparing and analyzing information processed for a document with related information received from an external server device, and generating analysis data for a document.

10 is a diagram for explaining a UI for providing analysis data for a document to a user when it is determined as a safe sale according to a document analysis method according to an embodiment of the present disclosure.

11 is a diagram for explaining a UI for providing analysis data for a document to a user when it is determined that a safety device is required for sale according to a document analysis method according to an embodiment of the present disclosure.

12 is a view for explaining a UI for providing analysis data for a document to a user when it is determined as a dangerous sale according to a document analysis method according to an embodiment of the present disclosure.

13 is a diagram for explaining a UI for providing analysis data for a document to a user when it is determined that there is an abnormality in the document according to the document analysis method according to an embodiment of the present disclosure.

14 is a diagram for explaining a UI for providing analysis data including detailed analysis results for a document to a user according to a document analysis method according to an embodiment of the present disclosure;

15 is a diagram for explaining a UI for providing analysis data including information related to a document to a user according to a document analysis method according to an embodiment of the present disclosure;

FIG. 16 is a diagram for explaining a UI for providing a user with information on the degree of risk related to whether a deposit can be fully received upon termination of a jeonse contract when the type of property for sale is a collective building.

FIG. 17 is a diagram for explaining a UI for providing a user with information on the degree of risk related to whether a deposit can be fully received upon termination of a jeonse contract when the type of sale is a multi-family house.

18 is a flowchart illustrating a document analysis method according to an embodiment of the present disclosure.

It should be understood that the embodiments described below are illustratively shown to help the understanding of the present disclosure, and the present disclosure may be implemented with various modifications, different from the embodiments described herein. However, in the following description of the present disclosure, if it is determined that a detailed description of a related well-known function or component may unnecessarily obscure the gist of the present disclosure, the detailed description and specific illustration thereof will be omitted. In addition, the accompanying drawings are not drawn to scale in order to help understanding of the disclosure, but dimensions of some components may be exaggerated.

The terms used in this specification and claims have been chosen in consideration of the function of the present disclosure. However, these terms may vary depending on the intention of a person skilled in the art, legal or technical interpretation, and emergence of new technology. Also, some terms are arbitrarily selected by the applicant. These terms may be interpreted in the meanings defined herein, and in the absence of specific definitions, they may be interpreted based on the general content of the present specification and common technical common sense in the art.

In the description of the present disclosure, the order of each step should be understood as non-limiting unless the preceding step must be logically and temporally performed before the subsequent step. In other words, except for the above exceptional cases, even if the process described as the subsequent step is performed before the process described as the preceding step, the essence of the disclosure is not affected, and the scope of rights should also be defined regardless of the order of the steps.

In this specification, expressions such as “have,” “may have,” “include,” or “may include” indicate the presence of a corresponding characteristic (eg, a numerical value, function, operation, or component such as a part). and does not exclude the presence of additional features.

Terms such as first, second, etc. may be used to describe various elements, but the elements should not be limited by the terms. The above terms may be used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present disclosure, a first component may be referred to as a second component, and similarly, a second component may also be referred to as a first component.

In addition, since the present specification describes components necessary for the description of each embodiment of the present disclosure, the present disclosure is not necessarily limited thereto. Accordingly, some components may be changed or omitted, and other components may be added. In addition, they may be distributed and arranged in different independent devices.

Further, an embodiment of the present disclosure will be described in detail below with reference to the accompanying drawings and the contents described in the accompanying drawings, but the present disclosure is not limited or limited by the embodiments.

Hereinafter, the present disclosure will be described in detail with reference to FIGS. 1 to 16 .

The server device 100 may communicate with external devices. For example, referring to FIG. 1 , the server device 100 may communicate with the user terminal device 200 , the external server device 300 , and the third-party terminal device 400 . However, the number of terminals capable of communicating with the server device 100 is not limited thereto, and the type or communication method of the target terminal to communicate with may be implemented in various ways.

A detailed description related to the operation of the server device 100 will be described later with reference to FIG. 2 .

The user terminal device 200 is a terminal device used by a user who wants to receive an analysis result for a document. The user may provide the document 10 (refer to FIG. 3 ) to be analyzed to the server device 100 through the user terminal device 200 . The user provides the document 10 stored in the memory 220 of the user terminal device 200 to the server device 100 or connects the UBS memory in which the document 10 is stored to the user terminal device 200 to the user terminal device It may be provided to the server device 100 through 200 . Here, the document 10 may be a word file or an image file.

On the other hand, the user can provide the document 10 file to the server device 100 through the user terminal device 200 as well as directly input related information into the user terminal device 200 to the server device 100 can provide For example, the user may input information in text form into the user terminal device 200 and provide it to the server device 100 .

The user terminal device 200 may use various types of devices such as a PC, a laptop PC, a mobile phone, a tablet PC, etc., but is not limited thereto, and the user stores the file of the document 10 in a USB memory, The document 10 may be provided to the server device 100 by connecting to 100 ) and directly inputting the document 10 .

The external server device 300 may communicate with the server device 100 . For example, the server device 100 may request the external server device 300 for related information corresponding to information recognized from the document 10 to receive the corresponding information from the external server device 300 .

The server device 100 may provide a document analysis result to the third party terminal device 400 , and may reflect information received from the third party terminal device 400 in document analysis. Also, the server device 100 may provide information received from the third party terminal device 400 to the user terminal device 200 .

For example, the server device 100 may be a device for analyzing real estate sales and real estate contract documents and providing an analysis result to a user, and the user terminal device 200 is a terminal used by a user who wants to receive an analysis service It is a device, and the third-party terminal device 400 may be a terminal device used by a real estate agent or an individual who provides real estate sale information to the server device 100 . Also, the external server device 300 may be an external server including information related to real estate specified by a real estate contract, for example, a real estate register copy, land ledger, building ledger, and the like.

In this case, the server device 100 may analyze the real estate for sale information provided from the third party terminal device 400 and provide the analysis result to the user terminal device 200 . When the user provides the real estate contract document to the server device 100 through the user terminal device 200 , the server device 100 may analyze the real estate contract document and provide the analysis result to the user terminal device 200 . Also, information may be transmitted/received between the user terminal device 200 and the third-party terminal device 400 through the server device 100 .

Hereinafter, the configuration and operation of the server device 100 according to an embodiment of the present disclosure will be described with reference to FIG. 2 .

Referring to FIG. 2 , the server device 100 according to an embodiment of the present disclosure may include a communication unit 110 , a memory 120 , and a processor 130 .

The communication unit 110 may transmit/receive various data by performing communication with an external device. For example, the communication unit 110 may receive a document and related information from the user terminal device 200 , the external server device 300 , and the third party terminal device 400 , and transmit analysis data for the document to the user terminal may be transmitted to the device 200 .

Meanwhile, the communication unit 110 may communicate with an external device through a local area network (LAN), an Internet network, and a mobile communication network, as well as BT (Bluetooth), BLE (Bluetooth Low Energy), WI - It is possible to communicate with an external device through various communication methods such as FI (Wireless Fidelity), WI-FI Direct (Wireless Fidelity Direct), Zigbee, NFC, etc. To this end, the communication unit 110 may include various communication modules for performing network communication. For example, the communication unit 110 may include a Bluetooth chip, a Wi-Fi chip, a wireless communication chip, and the like.

The memory 120 may store an operating system (OS) for controlling overall operations of the components of the server device 100 and commands or data related to components of the server device 100 .

For example, the memory 120 may store a file for a document received from the user terminal device 200 . Also, the memory 120 may store document related information received from the external server device 300 .

The processor 130 may control a plurality of hardware or software components of the server device 100 using various commands or data stored in the memory 120 , and a command received from at least one of the other components or Data may be loaded into a volatile memory for processing, and various data may be stored in a non-volatile memory.

The processor 130 according to an embodiment of the present disclosure receives the document 10 stored in the memory 120 or receives the document 10 received from the external device through the communication unit 110, and receives the inputted document ( 10) may be extracted, texts corresponding to a plurality of identification items may be identified among the extracted texts, and a text written in an area adjacent to the identified text may be recognized as information corresponding to the identification items.

In addition, the processor 130 classifies the information corresponding to the identification item into a plurality of preset groups, processes information belonging to the plurality of groups according to a processing process set differently for each group, and processes the information belonging to the plurality of groups based on the processing result. ) can provide analysis data for A detailed description related thereto will be provided later.

Referring to FIG. 3 , the user terminal device 200 according to an embodiment of the present disclosure may include a communication unit 210 , a memory 220 , a display 230 , and a processor 240 .

The communication unit 210 may transmit/receive various data by performing communication with an external device, for example, the server device 100 or the third-party terminal device 400 . For example, the communication unit 210 may receive a document and related information from the server device 100 and the third-party terminal device 400 , and may receive analysis data for the document from the server device 100 . .

The memory 220 may store an operating system (OS) for controlling overall operations of components of the user terminal device 200 and commands or data related to components of the user terminal device 200 . For example, the memory 220 may store a document 10 file, and the document 10 may be a word file or an image file captured by a camera (not shown) included in the user terminal device 200 . .

The display 230 is a configuration for displaying an image, and includes a liquid crystal display (LCD), an organic light-emitting diode (OLED), a liquid crystal on silicon (LCoS), a digital light processing (DLP), and a quantum dot (QD) display. It may be implemented in various forms, such as a panel.

In this case, the display 230 may display a UI for providing information to the user. For example, the display 230 may display a UI that provides analysis data for a document to a user according to the document analysis method according to an embodiment of the present disclosure. A detailed description of the UI for providing analysis data for a document will be described later with reference to FIGS. 10 to 15 .

The processor 240 may control a plurality of hardware or software components of the user terminal device 200 using various commands or data stored in the memory 220 , and a command received from at least one of the other components. Alternatively, data may be loaded into a volatile memory for processing, and various data may be stored in a non-volatile memory.

For example, the processor 240 may control the communication unit 210 to transmit/receive signals to and from an external device, and may control the display 230 to display a UI providing analysis data for a document.

Referring to FIG. 4 , a document 10 to be analyzed according to an embodiment of the present disclosure may be a real estate contract document. However, the document to be analyzed according to the present disclosure is not limited thereto, and may be applied to various types of documents.

Hereinafter, for convenience of description, the document 10 to be analyzed according to an embodiment of the present disclosure is a real estate contract document.

Typically, a document may be composed of a layout and content. The layout may include various types of information constituting the form of a document. Content means meaningful information filled in the layout. For example, when the document includes information in the form of a table, the layout may correspond to lines constituting the table, and the content may be various texts filled in the table.

However, the layout is not limited thereto, and not only a form such as a table including various lines, but also a layout in which contents are arranged in a certain form even if the line is not included may also correspond to the layout.

When a layout and text are separated in the document 10 of FIG. 4 , lines constituting a table included in the document 10 may be classified as a layout. In addition, a form in which the texts 12-8, 12-9, and 12-10 arranged at regular intervals up and down may also correspond to a layout.

Meanwhile, texts 11 and 12 written inside each cell included in the table layout may be classified as content.

The server device 100 may separate the layout and text included in the document 10 and process them in different ways.

Separation of layout and text can be accomplished in a variety of ways.

Hereinafter, a case in which the document 10 includes a table layout among methods of extracting a layout and text from the document 10 will be described as an example with reference to FIGS. 5 to 7 .

5 is a diagram for explaining an operation of extracting a vertical line from a table layout included in a document, FIG. 6 is a diagram for explaining an operation of extracting a horizontal line from a table layout included in a document, and FIG. A diagram for explaining an operation of identifying a cell included in a table layout by obtaining an intersection between an extracted vertical line and a horizontal line.

5 and 6 , when the document 10 includes a table layout, the processor 130 of the server device 100 performs vertical lines 13-1, 13-2, 13-3, and 13 in the table layout. -4, 13-5, 13-6, 13-7, 13-8, see Fig. 5) and horizontal lines (14-1, 14-2, 14-3, 14-4, 14-5, 14-6, 6) can be extracted, respectively. Erosion and dilation operations may be used as a method of extracting the vertical and horizontal lines of the table layout, but are not limited thereto. Also, if the interval between components included in the layout is constant, the margin may be recognized as a boundary line.

Referring to FIG. 7 , the processor 130 of the server device 100 merges the extracted vertical and horizontal lines to obtain intersections between the vertical and horizontal lines, and to identify cells included in the table layout based on the obtained intersections. can For example, when four different intersecting points are arranged to form a single rectangle as each vertex, the server device 100 may identify that one cell is located in an area formed by the corresponding intersecting points.

The processor 130 of the server device 100 may identify a cell included in the table layout and extract an object located inside the identified cell as text. In this case, the server device 100 may extract and classify text by applying a Deep Learning OCR (Optical Character Recognition) model to an object located inside the cell. For example, when classification is performed by applying the OCR model to a cell containing the text “Location” in FIG. 7, “so”, “jae”, and “ji” are recognized and classified as each syllable, and the three syllables are Since it is located inside one cell, each syllable can be merged into one and completed as the text “Location”.

Meanwhile, the processor 130 may determine whether an object disposed inside a cell constitutes text. The processor 130 may determine whether or not the text corresponds to the text by comparing the shape of the characters described in the image DB and the image DB of various languages such as Korean, English, Chinese, and Japanese, and the shape of an object disposed inside the cell.

Also, if it is determined that a plurality of texts are arranged in a uniformly arranged form, the processor 130 may determine that the corresponding texts form a uniform layout.

Meanwhile, the processor 130 may recognize texts corresponding to a plurality of identification items and text corresponding to information corresponding to the identification items from among the texts extracted from the document 10 .

Here, the identification item means an item for identifying each type of various information included in the document 10 .

For example, the identification item may be an item for identifying a plurality of features included in information on real estate to be contracted in a real estate contract document. Referring to FIG. 4 , identification items such as 'location' for specifying real estate may be described in the document 10, and 'land', 'location', 'area', ' Identification items such as 'building', 'structure', and 'use' and identification items such as 'special agreement' including information on the contents of a real estate contract may be described.

However, the plurality of identification items is not limited thereto, and addresses (eg, “location” indicating the real estate sale address, address related to the contractor’s personal information, etc.), floor number, lessor, owner, land area, structure, use, building It can include various items such as area, deposit, purchase price, and special contract.

The processor 130 may identify a text disposed at a specific location as an identification item or determine whether it is an identification item based on the content of the text. For example, in a tabular layout, the text written inside the cell corresponding to column 1 is identified as the text corresponding to the identification item, or 'lessor', 'tenant', 'location', 'land', 'position', ' By recognizing words that are normally designated as identification items in real estate contract documents, such as 'area', 'building', 'structure', 'use', 'special terms', etc., it can be determined whether the text corresponds to the identification item.

In this case, the information corresponding to each identification item may be information indicated by text written in an area adjacent to the text corresponding to the identification item. For example, in the processor 130, the information corresponding to the 'location' 11-1 of the real estate is '1234-12 Bongcheon-dong, Gwanak-gu, Seoul' (12-1), and the 'building' (11-5) ' Information corresponding to the 'structure' 11-6 can be recognized as 'reinforced concrete' (12-4).

In addition, the processor 130 records the texts (12-8, 12-9, 12-10) arranged at regular intervals up and down in the area adjacent to the 'special agreement' 11-10 to the 'special agreement' (11-). It can be recognized as information corresponding to 10).

Meanwhile, the processor 130 may classify information corresponding to the identification item into a plurality of preset groups, and process the information for each group.

Hereinafter, an operation in which the processor 130 classifies information corresponding to an identification item into a plurality of groups and processes the information for each group will be described with reference to FIG. 8 .

8 is a diagram illustrating data obtained for a document in a table form. Referring to FIG. 8 , the processor 130 may acquire data about the document 10 by recognizing the information 22 corresponding to each identification item 21 .

The processor 130 classifies the information 22 corresponding to the identification item 21 into a plurality of groups based on the obtained data in this way, and collects information belonging to the plurality of groups according to a processing process set differently for each group. can be processed

For example, the processor 130 is divided into a first group including information on contract details of real estate, a second group including information indicating characteristics of real estate, and a third group including information for specifying real estate. The identification item 21 and the information 22 corresponding to the identification item may be classified.

Referring to FIG. 4 , among the identification items, the 'special contract' 21 - 8 is information on the contract details of real estate, and thus may be classified into the first group. In addition, among the identification items, 'structural use' (21-5), 'building area' (21-6), 'deposit' (21-7), etc. are information indicating the characteristics of real estate, so they can be classified into the second group. , 'address' (21-1), 'floor number' (21-1), 'lessor' (21-3), 'account owner' (21-4), etc. are information for specifying real estate, so the third group can be classified as

The processor 130 may apply and process different processing processes to information belonging to each group according to the characteristics of the plurality of groups. In this case, the plurality of groups may be classified based on information required for each processing process.

For example, since the first group includes information 22 - 8 on contract details of real estate, an in-depth analysis may be required. Accordingly, the processor 130 processes the information belonging to the first group by applying the first processing process including spelling correction, spacing correction, number conversion, term conversion, related content addition, and property safety determination process to the first group. can do.

Here, the number conversion may mean converting text with numbers and letters mixed into numbers only (for example, converting '1 million won' to '100000') when expressing the amount, and the term conversion is in the real estate contract document. It may mean converting contained legal terms into ordinary terms.

Adding related content may mean adding real estate-related legal knowledge, general real estate information, and customs that may be unfamiliar to general users in the information processing process. For example, adding information that there may be difficulties in getting a cheonsei loan if a high mortgage is set on a real estate sale, or in the case of a transaction that meets certain requirements (for example, if the owner is a corporation or the type of building is a neighborhood living facility) Information that it may be difficult to subscribe to Jeonse Guarantee Insurance can be added as related content provided to users.

In addition, the property stability determination process may refer to information processing, such as calculating the safety level of the deposit price, for example.

The second group includes information 22-5, 22-6, and 22-7 indicating the characteristics of real estate, and it is possible to provide necessary information to the user even though a lower level analysis than the first group is performed. Accordingly, the processor 130 applies a second processing process including spacing correction, number conversion, term conversion, accredited information match determination process, property safety determination process, and property suitability determination process based on user preference information to the second group Thus, information belonging to the second group can be processed.

Here, the process for determining suitability of a property based on user preference information may refer to information processing for determining whether a property is suitable for sale according to user preference information input from a user.

The third group includes information (22-1, 22-2, 22-3, and 22-4) for specifying real estate, and the third group also performs a lower level analysis than the first group. information can be provided. Accordingly, the processor 130 may process the information belonging to the third group by applying a third processing process including spacing correction, term conversion, a process for determining whether public information matches, and a process for determining safety for sale to the third group. .

Accordingly, by applying a different level of information processing process to each group, it is possible to efficiently process information included in the document 10 .

The processor 130 may process information belonging to each group according to a processing process set differently for each group, and provide analysis data for the document 10 based on the processing result.

Hereinafter, an operation of generating analysis data for the document 10 based on the information processing result will be described with reference to FIG. 9 .

The server device 100 may receive information related to the document 10 from the external server device 300 through the communication unit 110 . For example, the server device 100 is a certified copy of the register of real estate specified by the document 10 from the external server device 300, land ledger, building ledger, cadastral map, land use plan confirmation, real estate price information (market price information, It is possible to receive at least one piece of information among public price information, official land price information, etc.), real estate terminformation (information related to laws such as the Enforcement Decree of the Building Act, the Enforcement Decree of the Housing Act, terminology information of apartment houses, type 1 neighborhood living facilities, etc.).

In this case, the processor 130 may generate analysis data for the document 10 based on the related information received from the external server device 300 and information recognized as the document 10 .

Hereinafter, with reference to FIG. 9 , in various embodiments of the operation of generating analysis data, the corresponding information among the related information received from the external server device 300 and the information recognized from the document 10 is compared to determine whether they match. Describe the action to be performed.

Referring to FIG. 9 , the server device 100 may receive study document related information for real estate specified in the real estate contract document from the external server device 300 .

The processor 130 compares and analyzes the

data

21 and 22 obtained for the document 10 and the

data

31 and 32 for the study document, which is the related information received from the external server device 300 , and analyzes the data (40) may be provided to the user.

Specifically, the processor 130 matches the identification item 21 of the real estate contract corresponding to the document 10 to be analyzed and the identification item 31 of the study document to correspond, and corresponds to the identification item of the contract By comparing the information (22) and the information (32) corresponding to the identification item of the study document, it is possible to determine whether the match.

For example, in FIG. 5, the information corresponding to the account owner of the contract is 'Park○○', whereas the information corresponding to the owner described in the registered copy is 'Kim○○', so the processor 130 is in the contract. It is possible to generate analysis data (40-5) that the recorded information and the information described in the registered copy do not match.

In addition, referring to Figure 9, the processor 130, based on the information corresponding to the security deposit of the contract and the information corresponding to the root mortgage of the registered copy, it is possible to generate analysis data 40-8 for the safe range of the deposit, , data processing for this can be performed on the server.

For example, after analyzing statistical information such as distribution related to a specific area in a specific period, what percentage of the market price is the auction price of a specific property, and what percentage of the actual loan price for a recent near-equity mortgage, the safe range is based on the analyzed information can be calculated. For example, when analyzing a sale for an apartment in Gwanak-gu, Seoul, if the auction price is 80% of the average market price and the mortgage is 120% of the average actual loan price, the safe amount to be set for the real estate when making a contract for the real estate Since is an amount obtained by subtracting 10/12 times the mortgage amount from 0.8 times the market price of the apartment for sale, it is possible to generate the analysis data 40-8 for recommending a safety setting of a certain amount when the deposit amount is lower than the safe amount.

On the other hand, as another example of generating the analysis data, the processor 130 recognizes from the related information and document 10 received from the external server device 300 according to the property safety determination process that determines the property safety through a preset algorithm. The information can be processed to generate analytic data.

For example, when the processor 130 has a special word such as 'joint collateral' in the certified copy of the register received from the external server device 300, the published price and near-mortgage setting information of several items included in the joint collateral list of the certified copy of the register In consideration of this, it is possible to generate analysis data that comprehensively judges the safety of the product.

On the other hand, as another example of generating the analysis data, the processor 130 is the related information and documents received from the external server device 300 according to the sale suitability determination process for determining the suitability for sale based on the user preference information input from the user. Analysis data may be generated by processing the information recognized from (10).

As an example, the server device 100 receives user preference information on 'whether or not to subscribe to HUG Jeonse Guarantee Insurance' from the user, and in the related information received from the external server device 300, the purpose of the contracted real estate is 'neighborhood life'. In the case of 'facility', it is possible to generate analysis data including information that it is not covered by HUG Jeonse Guarantee Insurance.

Accordingly, the processor 130 may generate analysis data such as the accuracy of information related to the real estate and the owner specified in the real estate contract document, information on setting rights related to the real estate, and the safety of the real estate contract.

Analysis data for the document 10 generated as described above may be provided to the user. As an example, the analysis data may be provided to the user through a UI displayed on the user terminal device 200 .

Hereinafter, a UI displayed to provide the user with analysis data generated for the document 10 will be described with reference to FIGS. 10 to 15 .

10 to 15 are diagrams for explaining a UI for providing analysis data for a document to a user according to a document analysis method according to an embodiment of the present disclosure.

The server device 100 may transmit the analysis data generated by the processor 130 to the user terminal device 200 through the communication unit 110 . The user terminal device 200 may receive the analysis data and provide the analysis data to the user by displaying the UI on the display 230 .

FIG. 10 exemplarily shows a UI provided to a user when the document 10 is analyzed as a safe sale as the target of the contract as a result of analysis.

Referring to FIG. 10 , the UI provided to the user may include a UI 1010 for displaying a menu and a UI 1020 for displaying analysis data. The UI 1020 for displaying the analysis data may display information on the safety of the real estate contract that the contents of the listing of the sale subject to the contract are on the safe side.

Information on the safety of a real estate contract may include a value expressing the safety numerically. In this case, the information on the safety of the real estate contract may be provided in the form of at least one of text or numbers.

11 shows that the document 10 analysis result shows that the sale subject to the contract is a sale that requires a safety device, for example, as shown in FIG. The UI that is used is shown as an example.

Referring to FIG. 11 , the UI provided to the user may include a UI 1110 for displaying a menu and a UI 1120 for displaying analysis data. The UI 1120 for displaying the analysis data may display information on the safety of a real estate contract including information that a user can refer to in order to secure a safe right to a sale target of the contract.

12 shows that when the content of the contract is analyzed as dangerous as a result of the analysis of the document 10, for example, there is an abnormality in the special clause or there is an abnormality limiting the ownership in the registered copy of the real estate subject to the contract. When analyzed as such, the UI provided to the user is shown as an example.

Referring to FIG. 12 , the UI provided to the user may include a UI 1210 for displaying a menu and a UI 1220 for displaying analysis data. The UI 1220 for displaying the analysis data may display information on the safety of a real estate contract including information that a user can refer to in order to secure a safe right to a sale target of the contract.

Specifically, with reference to FIGS. 13 to 15 , when the risk is determined in the contract content, the user terminal device 200 analyzes data for a contract, a registered copy, etc. based on the analysis data received from the server device 100 . The UI to be displayed can be provided to the user, and detailed information about the disadvantages that may occur to the user and the rights set in the registered copy can be displayed.

As an example, when it is determined that the mortgage is at a risk level as a result of the analysis of the certified copy of the register, the user terminal device 200 may provide the user with a UI that displays analysis data related to whether the user can fully receive the deposit at the end of the jeonse contract. there is. In this case, analysis data analyzed in consideration of different legal considerations for each type of sale can be displayed through an intuitively schematic graph or text.

16 and 17 show a UI for providing a user with information on the degree of risk related to whether the deposit can be fully received upon termination of the jeonse contract when the type of sale is an aggregate building and a multi-family house, respectively.

Referring to FIG. 16 , the UI provided to the user is a UI 1610 indicating that the mortgage on the register is at a risk level, a UI 1620 indicating the level of the mortgage against the market price, and a solution that can be requested from the lessor if the mortgage is at a risk level It may include a UI 1630 for presenting , a UI 1640 for selecting an amount for requesting cancellation of a mortgage, and a UI 1650 for receiving an input for adding a special contract.

For example, if the analysis result determines that the near-mortgage set relative to the market price of the property is at a risky level, it may be suggested to the lessor to request such as a request to cancel the near-mortgage from the lessor or to cancel the near-mortgage with the user's deposit.

In this case, the UI provided to the user may present the recommended amount to cancel the mortgage. As an example, the server device 100 determines what percentage of the market price is the auction price of a specific area and a specific item for a specific recent period, based on data obtained by performing statistical processing on information received from the external server 300 , recently By analyzing statistical information such as distribution related to the percentage of the actual loan price, the recommended amount to be canceled within a safe range can be calculated.

On the other hand, Figure 17 shows the UI provided when the sale is a multi-family house. Based on the information on the registered copy and other information (eg, senior rent, the number of small tenants in the multi-family building, etc.), the sale is auctioned If passed to , information on whether the user's deposit can be safely preserved can be displayed.

For example, it is possible to determine the risk level of the mortgage right set for the sale based on information such as the relative mortgage setting information on the registered copy, the amount of the senior rental amount, the number of small tenants, and the like, and provide it to the user.

Here, the above-described information may be received from the external server device 300 or from data input into the third-party terminal device 400 by a lessor or a real estate agent. It is a flowchart for explaining the document analysis method.

The server device may receive a document input (S1810).

Here, the document may be a document file created by various programs such as Doc, Hwp, Xls, etc., may be an image file such as PDF, or may be an image file obtained by converting a paper document provided offline using a camera or a scanner. .

Then, the server device may extract the text included in the document (S1820).

Then, the server device may identify the text corresponding to the identification item from among the extracted texts (S1830).

Then, the server device may recognize the text written in the area adjacent to the identified text as information corresponding to the identification item (S1840).

Then, the server device may classify the information corresponding to the identification item into a plurality of preset groups ( S1850 ).

In this case, the plurality of groups may include first to third groups, the first group includes information on real estate contract details, the second group includes information indicating real estate characteristics, and the third group includes It may include information for specifying real estate.

In addition, the server device may process information belonging to a plurality of groups according to a processing process set differently for each group ( S1660 ).

In this case, based on the information belonging to at least one of the plurality of groups, the related information is received from the external server, and the related information received from the external server is compared with the corresponding information among the information recognized from the document and matched. can determine whether

Then, the server device may provide analysis data for the document based on the processing result (S1670).

Meanwhile, the above-described methods according to various embodiments of the present disclosure may be implemented in the form of software or applications that can be installed in an electronic device such as a user terminal device.

In addition, the above-described methods according to various embodiments of the present disclosure may be implemented only by software upgrade or hardware upgrade of an existing electronic device.

In addition, various embodiments of the present disclosure described above may be performed through an embedded server provided in the electronic device or a server external to the electronic device.

Meanwhile, a non-transitory computer readable medium in which a program for sequentially performing the document analysis method according to the present disclosure is stored may be provided.

The non-transitory readable medium refers to a medium that stores data semi-permanently, rather than a medium that stores data for a short moment, such as a register, cache, memory, and the like, and can be read by a device. Specifically, the various applications or programs described above may be provided by being stored in a non-transitory readable medium such as a CD, DVD, hard disk, Blu-ray disk, USB, memory card, ROM, and the like.

In the above, preferred embodiments of the present disclosure have been illustrated and described, but the present disclosure is not limited to the specific embodiments described above, and it is common in the technical field to which the disclosure belongs without departing from the gist of the present disclosure as claimed in the claims. Various modifications may be made by those having the knowledge of

Claims

receiving a document input;

extracting text included in the document;

identifying texts corresponding to a plurality of identification items from among the extracted texts;

recognizing a text written in an area adjacent to the identified text as information corresponding to the identification item;

classifying the information corresponding to the identification item into a plurality of preset groups;

processing information belonging to the plurality of groups according to a processing process set differently for each group; and

A document analysis method comprising a; providing analysis data for the document based on the processing result.
According to claim 1,

The processing of information belonging to the plurality of groups includes:

receiving related information from an external server based on information belonging to at least one of the plurality of groups; and

Generating analysis data for the document based on the related information received from the external server and information recognized from the document;
3. The method of claim 2,

The step of generating the analysis data comprises:

Comparing the related information received from the external server and corresponding information among the information recognized from the document to determine whether or not they match; including, a document analysis method.
3. The method of claim 2,

The document is a real estate contract document,

The step of generating the analysis data comprises:

A document analysis method for generating the analysis data by processing the related information received from the external server and information recognized from the document according to a property safety determination process of determining property safety through a preset algorithm.
3. The method of claim 2,

The document is a real estate contract document,

The step of generating the analysis data comprises:

A document analysis method for generating the analysis data by processing related information received from the external server and information recognized from the document according to a property suitability determination process of determining property suitability based on user preference information input from a user.
According to claim 1,

The document is a real estate contract document,

The processing process is

A first processing process set to be applied to the first group among the plurality of groups, including spelling correction, spacing correction, number conversion, term conversion, related content addition, and property safety determination process;

A second process set to be applied to the second group among the plurality of groups, including space correction, number conversion, term conversion, accredited information match determination process, property safety determination process, and property suitability determination process based on user preference information process,

It includes a third processing process set to be applied to a third group among the plurality of groups, including spacing correction, term conversion, a process for determining whether official information matches or not, and a process for determining safety for sale,

The plurality of groups are classified based on information required for each processing process, a document analysis method.
7. The method of claim 6,

The first group includes information about the contents of the real estate contract,

The second group includes information indicative of real estate characteristics;

The third group includes information for specifying real estate, a document analysis method.
According to claim 1,

The document is a real estate contract document,

The analysis data for the document includes information on at least one of the accuracy of information related to the real estate and the owner specified in the real estate contract document, information on setting rights related to the real estate, and the safety of the real estate contract, a document analysis method .
9. The method of claim 8,

The information on the safety of the real estate contract is provided in the form of at least one of text, number, image, or video, a document analysis method.
3. The method of claim 2,

The document is a real estate contract document,

Receiving the relevant information from the external server comprises:

Receiving at least one information of a registered copy of real estate specified by the document, land ledger, building ledger, cadastral map, land use plan confirmation, real estate price information, and real estate term information from at least one external server,

The step of generating the analysis data comprises:

A document analysis method for generating analysis data based on data obtained by performing statistical processing on the information received from the external server.
According to claim 1,

The plurality of identification items include at least one of an address, a floor number, a lessor, an owner, a land area, a structure, a use, a building area, a deposit, a purchase price, and a special contract, a document analysis method.
3. The method of claim 2,

Receiving the relevant information from the external server comprises:

Receiving related information corresponding to each of a plurality of different viewpoints,

The step of providing the analysis data,

Based on the received related information, a document analysis method for generating and providing a plurality of analysis data corresponding to each of the plurality of time points.
13. The method of claim 12,

The document is a real estate contract document,

The plurality of time points include at least two of before the real estate contract date, the real estate contract date, and after the real estate contract date, a document analysis method.
According to claim 1,

The step of extracting the text is

when the document includes a table layout, extracting a vertical line and a horizontal line from the table layout, respectively;

obtaining an intersection point between the vertical line and the horizontal line; and

Including; identifying a cell included in the table layout based on the intersection point,

A document analysis method for extracting an object located inside the identified cell as text.
15. The method of claim 14,

The step of extracting the text is

A document analysis method for extracting text by applying an OCR model to an object located inside the identified cell.