KR20220042838A - Systems and Methods to Provide AI Answering Service for Legal Questions - Google Patents

Abstract

One embodiment of the present invention relates to a system for providing a scored artificial intelligence (AI) answering service for legal questions, which displays negative and positive scores to a user in a gauge type. According to the present invention, the system comprises: an inquiry terminal requesting legal advice on inquiry content; and an AI legal question/answer server collecting case precedent information and legal consultation information of lawyers, to which cases similar to the inquiry content were applied, from at least one external server, processing the collected information to build a database, analyzing positive and negative legal answers to the inquiry content on the basis of the built case precedent information and legal consultation information by using an AI machine learning algorithm, and providing as positive and negative scores. The query terminal executes a question/answer platform displaying the scores to the user in a gauge type.

Description

Systems and Methods to Provide AI Answering Service for Legal Questions

The present invention relates to a system and method for providing a numerical AI answer service to legal questions.

In the legal field, counseling is basically required, but at present, most of them provide various means or forms (or forms) to connect experts and counseling and proceed only in the form of a preliminary investigation.

This standardizes the consultation and processes it by checking various steps and conditions in the ARS method, so the speed of the service is very tedious and the service experience is poor, which is a very big obstacle to the automated processing of the service. However, if consultation is made possible through natural language processing, the precision of consultation will increase according to machine learning, and the revitalization of related industries will be possible without time and space limitations.

Accordingly, the present invention intends to disclose a system and method that can provide an answer to a user's legal question using an AI learning algorithm.

Unexamined Patent Publication No. 10-2019-0143583 (Title of the Invention: A method and computer-readable recording medium for providing an answer based on a hybrid metrological conversation flow model in which machine learning is applied to a conversation management technology by a conversation understanding AI system )

An object of the present invention is to provide a system and method for providing a numerical AI answering service to legal questions that can solve conventional problems.

A system for providing a numerical AI answer service to a legal question according to an embodiment of the present invention for solving the above problem is a query terminal requesting legal advice on the query content: and a case to which an event similar to the query content is applied After collecting case law information and legal consultation information of lawyers from at least one or more external servers, it is processed into a preset form and built into a database. After analyzing legal answers to negation using an artificial intelligence machine learning algorithm, it includes an AI legal/question answer server that provides numerical values of positive and negative values respectively, and the query terminal uses the numerical values in the form of gauges for users It is characterized by running a question-and-answer platform that is displayed to

In one embodiment, the AI legal/question answering server converts the case precedent information and legal consultation information collected from the at least one or more external servers into a question-and-answer format, and then changes the tendency of the answer to 'positive (actively)' and 'negative (negative)' is characterized in that it is classified into any one of 'negative' and constructed as the database.

In one embodiment, when the AI law/question answer server builds the question and answer into a database, at least one or more of raw data for the question and answer, a category for the question, a case number, and the grounding law It is characterized in that it is built to be included.

In one embodiment, the AI law/question answering server is characterized in that it recognizes the sense of the user's query content as either positive or negative.

In one embodiment, the AI law/question answer server calculates sentence vectors of a plurality of questions and answers constructed in a database using a sentence vector similarity algorithm, and then calculates the sentence vectors of the 10 sentences with the highest similarity between sentences (questions and answers). ), define the perceived similarity that humans feel for the selected sentences as '0' or '1', then learn the defined sentences using the natural language processing learning algorithm (BERT NLP), and use the learned data It is characterized in that learning data is derived by re-learning.

In one embodiment, the AI law/question answer server is characterized in that it uses a natural language processing algorithm to extract semantic analysis, negative or positive tone, and a preset category for the query content.

In an embodiment, the AI law/question answer server processes the sentence content (meaning) in the query content into multidimensional space vector coordinates using a natural language processing learning algorithm (BERT NLP).

In an embodiment, the AI legal/question answering server extracts case precedent information and legal consultation information including a category and multidimensional space vector coordinates similar to the category of the query content and multidimensional space vector coordinates from the database. .

In one embodiment, the AI legal/question answering server is characterized in that it provides the average value of the positive and negative categories (numeric values) of the extracted at least one or more case precedent information and legal consultation information as the numerical value.

In one embodiment, the AI legal/question answering server has a positive or negative numerical value among the average values of the positive and negative categories (numeric values) of the extracted at least one or more case precedent information and legal consultation information is more than a relatively preset value When calculated as , it is characterized in that the numerical value reflecting the weight is provided.

A method of providing a numerical AI answering service to a legal question according to an embodiment of the present invention for solving the above problem includes the steps of: requesting legal advice on the content of a query from a query terminal; And the AI law/question answer server collects case precedent information to which cases similar to the above query are applied and legal consultation information of lawyers from at least one or more external servers, then processes them into a preset form to build a database, Based on case precedent information and legal consultation information, after analyzing legal answers to the affirmative and negation of the query contents using an artificial intelligence machine learning algorithm, providing a numerical value, wherein the query terminal uses the numerical value Characterized in executing a question-and-answer platform that displays the value to the user in the form of a gauge.

By using the system and method for providing a numerical AI answer service to a legal question according to an embodiment of the present invention, the user can conveniently find out whether the user is curious about himself or whether he has legal knowledge or legal application to the incidents that have occurred to him or her. There is an advantage in that answers to the questions can be provided in the form of gauges containing positive and negative figures, and similar precedents and cases can be provided.

1 is a network configuration diagram of a system that provides a numerical AI answering service to legal questions according to an embodiment of the present invention.
FIG. 2 is a detailed configuration diagram of the AI law/question answer server shown in FIG. 1. Referring to FIG.
3 is an exemplary diagram illustrating a data construction form of the database shown in FIG. 1 .
4 to 6 are exemplary views for explaining a natural language processing learning algorithm (BERT NLP).
7 and 8 are exemplary diagrams for explaining the execution process of the query and similarity analysis/extraction unit shown in FIG. 2 .
FIG. 9 is an exemplary diagram of a gauge bar in which the numerical value provided by the result providing unit shown in FIG. 2 is reflected in the query terminal.
10 is a flowchart of a method of providing a numerical AI answering service to a legal question according to an embodiment of the present invention.
11 is a detailed flowchart of the process S720 illustrated in FIG. 10 .
12 illustrates an example computing environment in which one or more embodiments disclosed herein may be implemented.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those of ordinary skill in the art can easily implement them. However, the present invention may be embodied in several different forms and is not limited to the embodiments described herein. And in order to clearly explain the present invention in the drawings, parts irrelevant to the description are omitted, and similar reference numerals are attached to similar parts throughout the specification.

Throughout the specification, when a part is "connected" with another part, this includes not only the case of being "directly connected" but also the case of being "electrically connected" with another element interposed therebetween. . In addition, when a part "includes" a certain component, it means that other components may be further included, rather than excluding other components, unless otherwise stated, and one or more other features However, it is to be understood that the existence or addition of numbers, steps, operations, components, parts, or combinations thereof is not precluded in advance.

The terms "about", "substantially", etc. to the extent used throughout the specification are used in or close to the numerical value when manufacturing and material tolerances inherent in the stated meaning are presented, and are intended to enhance the understanding of the present invention. To help, precise or absolute figures are used to prevent unfair use by unconscionable infringers of the stated disclosure. As used throughout the specification of the present invention, the term "step of (to)" or "step of" does not mean "step for".

In this specification, a "part" includes a unit realized by hardware, a unit realized by software, and a unit realized using both. In addition, one unit may be implemented using two or more hardware, and two or more units may be implemented by one hardware. Meanwhile, '~ unit' is not limited to software or hardware, and '~ unit' may be configured to be in an addressable storage medium or to reproduce one or more processors. Thus, as an example, '~' denotes components such as software components, object-oriented software components, class components, and task components, and processes, functions, properties, and procedures. , subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays and variables. The functions provided in the components and '~ units' may be combined into a smaller number of components and '~ units' or further separated into additional components and '~ units'. In addition, components and '~ units' may be implemented to play one or more CPUs in a device or secure multimedia card.

Some of the operations or functions described as being performed by the terminal, apparatus, or device in the present specification may be performed instead of by a server connected to the terminal, apparatus, or device. Similarly, some of the operations or functions described as being performed by the server may also be performed in a terminal, apparatus, or device connected to the server.

In this specification, some of the operations or functions described as mapping or matching with the terminal means mapping or matching the terminal's unique number or personal identification information, which is the identification data of the terminal. can be interpreted as

Hereinafter, based on the accompanying drawings, a system and method for providing a numerical AI answering service to a legal question according to an embodiment of the present invention will be described in more detail.

1 is a network configuration diagram of a system that provides a numerical AI answering service to legal questions according to an embodiment of the present invention, and FIG. 2 is a detailed configuration diagram of the AI legal/question answering server shown in FIG. 3 is an exemplary diagram illustrating a data construction form of the database shown in FIG. 1 , FIGS. 4 to 6 are exemplary views for explaining a natural language processing learning algorithm (BERT NLP), and FIGS. 7 and 8 are FIG. It is an exemplary view for explaining the execution process of the query and similarity analysis/extraction unit shown in 2, and FIG. 9 is an exemplary view of a gauge bar in which the numerical value provided by the result providing unit shown in FIG. 2 is reflected in the query terminal. .

First, as shown in FIG. 1 , the system 100 for providing a numerical AI answer service to a legal question according to an embodiment of the present invention includes a query terminal 200 and an AI legal/question answer server 300 . includes

However, since the system 100 for providing a numerical AI answer service to a legal question of FIG. 1 is only an embodiment of the present invention, the present invention is not limitedly interpreted through FIG. 1 .

At this time, each component of FIG. 1 is generally connected through a network. For example, as shown in FIG. 1 , at least one query terminal 200 may be connected to the AI law/question answer server 300 through a network.

Here, the network refers to a connection structure in which information exchange is possible between each node, such as a plurality of terminals and servers, and examples of such networks include RF, 3rd Generation Partnership Project (3GPP) network, and Long Term (LTE). Evolution) network, 5th Generation Partnership Project (5GPP) network, WIMAX (World Interoperability for Microwave Access) network, Internet, LAN (Local Area Network), Wireless LAN (Wireless Local Area Network), WAN (Wide Area Network) , PAN (Personal Area Network), Bluetooth (Bluetooth) network, NFC network, satellite broadcasting network, analog broadcasting network, DMB (Digital Multimedia Broadcasting) network, etc. include, but are not limited thereto.

In the following, the term at least one is defined as a term including the singular and the plural, and even if the at least one term does not exist, each element may exist in the singular or plural, and may mean the singular or plural. it will be self-evident In addition, that each component is provided in singular or plural may be changed according to embodiments.

The at least one query terminal 200 may be a terminal for inputting a search keyword and receiving a search result using a web page, an app page, a program, or an application related to a numerical AI answering service to a legal question. In addition, the at least one query terminal 200 may be a terminal that receives positive and negative answers to the search results as numerical information and then displays them in the form of a gauge.

Here, the at least one query terminal 200 may be implemented as a computer capable of accessing a remote server or terminal through a network. Here, the computer may include, for example, navigation, a laptop equipped with a web browser, a desktop, and a laptop. In this case, the at least one user terminal 100 may be implemented as a terminal capable of accessing a remote server or terminal through a network. At least one user terminal 100 is, for example, as a wireless communication device that guarantees portability and mobility, navigation, PCS (Personal Communication System), GSM (Global System for Mobile communications), PDC (Personal Digital Cellular), PHS(Personal Handyphone System), PDA(Personal Digital Assistant), IMT(International Mobile Telecommunication)-2000, CDMA(Code Division Multiple Access)-2000, W-CDMA(W-Code Division Multiple Access), Wibro(Wireless Broadband Internet) ) terminal, a smart phone, a smart pad, a tablet PC, etc. may include all types of handheld-based wireless communication devices.

Next, the AI law/question answer server 300 collects case precedent information to which a case similar to the above query is applied and legal consultation information of lawyers from at least one external server, and then processes it in a preset form to the database Based on the established case precedent information and legal consultation information, the legal answers to the positive and negative answers to the above questions are analyzed using an artificial intelligence machine learning algorithm, and then provided as numerical values of positive and negative values respectively. It may be a configuration that

The AI legal/question answer server 300 may be a server that provides a numerical AI answer service web page, app page, program or application to legal questions. AI law / question and answer server 300 detects the form of the query content, keywords of words or sentences included in the query content, and extracts a preset number of case law information and legal consultation information with high correlation or similarity with the keyword. After that, it may be of a configuration in which the extracted judgment information and the answer results (positive and negative) of the legal consultation information are provided as numerical information.

Here, the AI law/question answer server 300 may be implemented as a computer that can connect to a remote server or terminal through a network. Here, the computer may include, for example, navigation, a laptop equipped with a web browser, a desktop, and a laptop.

Referring to FIG. 2 , the AI law/question answer server 300 includes a collection unit 310 , a preprocessor 320 , a database conversion unit 330 , a machine learning unit 340 , a natural language processing unit 350 , and a query and similarity level. It includes an analysis/extraction unit 360 , a positive/negative quantification unit 370 , and a result value providing unit 380 .

The AI law/question answer server 300 or another server (not shown) operating in conjunction with at least one query terminal 200 and at least one information providing server (not shown) according to an embodiment of the present invention When transmitting a numerical AI answer service application, program, app page, web page, etc. to legal questions, at least one query terminal 200 installs or opens the corresponding service application, program, app page, web page, etc. can In addition, the service program may be driven in at least one query terminal 200 using a script executed in a web browser.

Here, the web browser is a program that enables the use of a web (WWW: world wide web) service. It means a program that receives and displays hypertext written in HTML (hyper text mark-up language), for example, Netscape. , Explorer, Chrome, etc. In addition, the application means an application on the terminal, for example, includes an app (app) executed in a mobile terminal (smartphone).

Referring to FIG. 2 , the collection unit 310 may be configured to collect precedent raw data and legal consultation (Q&A) data from at least one or more external servers.

Here, the external server may be a server of the National Penalty Information Center ( www.law.go.kr ) of the Republic of Korea, and the raw precedent data may be precedent big data of the district court, the high court, and the Supreme Court. In addition, legal consultation (Q&A) information includes legal information based on lawyers' answers to cases and answers.

The collection unit 310 may collect case law information and legal consultation information using the Selenium algorithm. In this case, the collection unit 310 may apply any program that can be collected by scraping case law information and legal consultation information.

The pre-processing unit 320 converts raw data and legal consultation data of case precedent information into an index, question, answer, category, case number, ground law, precedent result and consultation answer into affirmative/negative categories It can be a process of sorting/sorting/editing.

Here, the pre-processing unit 320 may transform the case precedent data into the form of a question (Q) and an answer (A) using a natural language processing algorithm.

For example, general case data is composed of many parts, such as <judgment matters>, <judgment summary>, <reference articles>, and <reference precedents>, and the contents are also complex. Therefore, in the present invention, only necessary parts of case data are extracted and pre-processed (categorized) so that similarity analysis and numerical answers can be provided to users in the future.

That is, the pre-processing unit 320 classifies the wording of the judgment into 'positive' and 'negative' based on the content of the issue of <judgment matters> among the case data, and quantifies it.

For example, in the case data provided by the National Legal Information Center, there is a consistent item called <judgment matter>, and issues in the form of [1] and [2] appear under the item.

For example, the numbers from <Judgment Matters> to [3] mean that there are three issues in the case. In each number you can see what the issue is and the judge's decision on that issue, marked as positive (“positive”) and negative (“passive”).

Next, the legal database conversion unit 330 classifies/sorts/categorizes the indexes, questions, answers, and tagging categories for questions preprocessed in the preprocessing unit, case numbers, ground law, case results, and affirmative/negative tagging of counseling answers into categories It may be a configuration that you edit and build into the database.

Here, the index may be a unique number (unique address) assigned to data registered in the database.

A question is the raw data of a natural language question. The answer may be an answer provided by a lawyer to the user's question.

The tagging category is an associative keyword related to a question, and may be, for example, lease, retirement, dismissal, and the like. The grounding law may be the grounding law for the attorney's answer to the case.

Negative/positive tagging is tagging information that determines whether the user's question form ends in an affirmative or negative form.

The negative/positive tagging is applied to the numerical value to be provided to the user. For example, if the question is 'Is it defamation?', it's positive, and if it's 'Isn't it defamatory?', it's negative. This is because the meaning of 20% of positive questions and 20% of negative questions in numerical answers is different in the case of such tagging.

Therefore, in the case of substituting an affirmative form for a numerical answer of a negative question later, (100 - Numerical answer) = This may be information to enable a substituted numerical answer.

Next, the machine learning unit 340 calculates sentence vectors of a plurality of questions and answers constructed in the database using the sentence vector similarity algorithm, and then selects 10 sentences (questions and answers) having the highest similarity between sentences. After defining the similarity of human feeling for the selected sentences as '0' or '1', learning the defined sentences using the natural language processing learning algorithm (BERT NLP), and re-learning the learned data. It performs the function of deriving learning data.

That is, the machine learning unit 340 may obtain a similarity between embedded sentence vectors based on cosine similarity and Jaccard similarity. In addition, 10 sentences having the highest similarity between sentences may be selected, but the present invention is not limited thereto.

In addition, data for learning BERT (Bidirectional Encoder Representations from Transformers) NLP (Natural Language Processing) by directly tagging 0 (similarity) and 1 (dissimilarity) for the degree of similarity felt by real humans to the selected 10 sentences After creating , it operates to learn BERT NLP.

For reference, BERT is designed to eliminate ambiguity by allowing the same word to have different embeddings depending on the sentence shape and position. For example, the banks of “account” and “of the river” have the same vector in Word2Vec or GloVe. However, this word should have a completely different meaning depending on the context, and in fact it has a completely different vector in ELMo or BERT. The example shown in the drawing is 1) I love you 2) I loved you 3) Love Who said it? If the cosine distance was checked with the result of morphological analysis of these three sentences as an input value, the word embedding value for the 'love' position was equally taken and the cosine distance was outputted. Among them, the 'love' vectors of Nos. 1 and 2, which have similar sentences, appear very close. On the other hand, the three sentences with completely different sentences are displayed quite far apart even if they are the same 'love'.

As another example, BERT may provide a similarity as a vector distance through cosine similarity between sentence or word vectors.

For example, when measuring the distance to 'love' vector 1 with the simple Sum of 'love' vectors 2 and 3, the distance between the two vectors Sum is about 0.17. Among them, we examine the distance from the weighted sum to which a weight is given to the second vector, which is close to each other. The distance was much closer. Conversely, we give weight to vector 3, which is farther away. You can see it goes further. Since cosine distance measures an angle, the distance is the same even if the vector increases, and the following shows the case where the vector shows the same distance even if the vector is 10 times larger. More details can be found in 　The meaning of cosine similarity.

Next, the natural language processing unit 350 may be configured to extract a semantic analysis, negative or positive tone, and a preset category for the query content using a natural language processing algorithm.

In analyzing the question content, the natural language processing unit 350 performs a semantic analysis vectorization process, negative/positive tagging process, and category tagging process of the corresponding question.

Here, the semantic analysis vectorization process may be a process of processing the query content with vector coordinates in a multidimensional space grasped based on the meaning of the question. In this case, natural language processing algorithms such as Universal Sentence Encoder and BERT are used. This allows questions with similar meanings to have unique coordinates that are close to each other.

Next, the negative/positive tagging process is a process of tagging by determining whether the user's question ends in an affirmative or negative form.

For example, if the question is 'Is it defamation?', it's positive, and if it's 'Isn't it defamatory?', it's negative. This is because the meaning of 20% of positive questions and 20% of negative questions in numerical answers is different in the case of such tagging. Therefore, in the case of substituting an affirmative form for a numerical answer to a negative question later, (100 - Numerical answer) = a substituted numerical answer.

Next, the category tagging process is a process of tagging keywords related to the content of the question, for example, it may be a process of tagging keywords for lease, retirement, dismissal, etc., which are field keywords or core keywords for the question content.

Next, the query and similarity analysis/extraction unit 360 may be configured to compare and analyze categories and multidimensional space vector coordinates similar to the category and multidimensional space vector coordinates of the query content with information in the database.

That is, the query and similarity analysis/extraction unit 360 first extracts case law information and legal consultation information having a category similar to the category of the query content from the database, and is similar to the multidimensional space vector coordinate of the query content (the coordinate distance is the most Near) information having multidimensional space vector coordinates may be secondarily extracted from information extracted first.

Meanwhile, the query and similarity analysis/extraction unit 360 may extract a legal case related to the query content through the above-described process.

The positive/negative quantification unit 370 calculates the average value of positive and negative categories (numeric values) for each answer of at least one case precedent information and legal consultation information extracted from the query and similarity analysis/extraction unit 360 It can be configuration.

The positive/negative numerical value unit 370 is a positive or negative numerical value among the average values of positive and negative categories (numeric values) of at least one or more extracted case precedent information and legal consultation information. , provides a numerical value reflecting the weight.

The result value providing unit 380 provides the positive/negative numerical values calculated by the positive/negative numerical value unit 370 in the form of a gauge graph.

The result providing unit 380 may use numerical values of several cases as input values and provide them through a nonlinear model similar to the method of using stacking in final decision making in the ensemble technique.

10 is a flowchart of a method of providing a numerical AI answer service to a legal question according to an embodiment of the present invention, and FIG. 11 is a detailed flowchart of the process S720 shown in FIG.

A method of providing a numerical AI answer service to a legal question shown in FIG. 10 may be implemented through the system shown in FIG. 1 .

10 and 11 , the method ( S700 ) of providing a numerical AI answer service to a legal question according to an embodiment of the present invention is first, a numerical AI answer to a legal question in the query terminal 200 . Requests legal advice or legal answers to the query contents to the AI legal/question answer server (S710) using a service-related web page, app page, program or application.

Here, the numerical value of the AI answer service related web page, app page, program or application to the legal question executed in the query terminal 200 is a gauge form can be displayed to the user.

When a query is received, the AI law/question answer server 300 collects case precedent information to which a legal case similar to the above query is applied and legal consultation information of lawyers from at least one external server, and then in a preset form It is processed and built into a database, and legal answers to the affirmative and negative answers to the above questions are analyzed using artificial intelligence machine learning algorithms based on the established case precedent information and legal consultation information, and then provided as numerical values ( S720).

More specifically, the S720 process converts the case precedent information and legal consultation information collected from the at least one or more external servers into a question-and-answer format, and then changes the tendency of the answer to 'positive (active)' and 'negative (passive)' )' may include the process of classifying it into any one of ' and building the database.

In addition, when constructing the question and answer as a database, it may include a process of constructing to include at least one of raw data for the question and answer, a category for the question, a case number, and a grounding law.

In addition, it may include a process of determining the sense of the user's query as any one of positive and negative.

In addition, after calculating sentence vectors of a plurality of questions and answers constructed in the database using the sentence vector similarity algorithm, 10 sentences (questions and answers) with the highest similarity between sentences are selected, and then, After defining the perceived similarity felt by humans as '0' or '1', learning the defined sentences using the natural language processing learning algorithm (BERT NLP), re-learning the learned data to derive learning data. can do.

In addition, it may include a process of extracting a semantic analysis, negative or positive tone of the query content, and a preset category using a natural language processing learning algorithm (BERT NLP).

In addition, the process may include processing the sentence content (meaning) in the query content into multidimensional space vector coordinates using the natural language processing learning algorithm (BERT NLP).

In addition, it may include a process of extracting from the database case precedent information and legal consultation information including a category and multidimensional space vector coordinates similar to the category of the query content and multidimensional space vector coordinates.

In addition, it may include the process of providing the average value of the positive and negative categories (numeric values) of the extracted at least one or more case precedent information and legal consultation information as the numerical value.

In addition, if the positive or negative numerical value among the average values of the positive and negative categories (numeric values) of at least one or more case precedent information and legal consultation information extracted is calculated to be more than a relatively preset number, a numerical value reflecting the weight is provided The process may further include

More specifically, the S720 process is a process executed by the AI legal/question answer server 300, and the collection unit 310 collects precedent raw data and legal consultation (Q&A) data from at least one external server in Korea. It is collected from an external server such as the server of the National Penalty Information Center ( www.law.go.kr ). In this case, raw case data may be case data of the district court, the high court, and the Supreme Court. In addition, legal consultation (Q&A) information includes legal information based on lawyers' answers to cases and answers.

Then, in the preprocessor 320, the raw data and legal consultation data of case precedent information are indexed, questions, answers, categories, case numbers, ground laws, case results and counseling answers are positive/negative categories. It carries out the process of classifying/sorting/editing the data and converting and processing the case precedent data into the form of a question (Q) and an answer (A) using a natural language processing algorithm.

When the operation of the preprocessor 320 is completed, the machine learning unit 340 calculates the sentence vectors of a plurality of questions and answers constructed in the database using the sentence vector similarity algorithm, After selecting a sentence (question and answer), defining the similarity of feeling that humans feel to the selected sentences as '0' or '1', learning the defined sentences using the natural language processing learning algorithm (BERT NLP), , it performs the function of deriving learning data by re-learning the learned data.

That is, the machine learning unit 340 may obtain a similarity between embedded sentence vectors based on cosine similarity and Jaccard similarity. In addition, 10 sentences having the highest similarity between sentences may be selected, but the present invention is not limited thereto.

In addition, data for learning BERT (Bidirectional Encoder Representations from Transformers) NLP (Natural Language Processing) by directly tagging 0 (similarity) and 1 (dissimilarity) for the degree of similarity felt by real humans to the selected 10 sentences After creating , it operates to learn BERT NLP.

Next, when the natural language processing unit 350 uses a natural language processing algorithm to analyze the meaning of the query content, to extract a negative or positive tone, to extract a preset category, and to analyze the question content, the semantic analysis vectorization process of the question , a negative/positive tagging process and a category tagging process are performed.

Here, the semantic analysis vectorization process may be a process of processing the query content with vector coordinates in a multidimensional space grasped based on the meaning of the question. In this case, natural language processing algorithms such as Universal Sentence Encoder and BERT are used. This allows questions with similar meanings to have unique coordinates that are close to each other.

Next, the negative/positive tagging process is a process of tagging by determining whether the user's question ends in an affirmative or negative form.

For example, if the question is 'Is it defamation?', it's positive, and if it's 'Isn't it defamatory?', it's negative. This is because the meaning of 20% of positive questions and 20% of negative questions in numerical answers is different in the case of such tagging. Therefore, in the case of substituting an affirmative form for a numerical answer to a negative question later, (100 - Numerical answer) = a substituted numerical answer.

Next, the category tagging process is a process of tagging keywords related to the content of the question, for example, it may be a process of tagging keywords for lease, retirement, dismissal, etc., which are field keywords or core keywords for the question content.

When the role of the natural language processing unit 350 is completed, the query and similarity analysis/extraction unit 360 compares and analyzes the category and multidimensional space vector coordinates similar to the category and multidimensional space vector coordinates of the query content with information in the database. do.

Here, the query and similarity analysis/extraction unit 360 primarily extracts case law information and legal consultation information having a category similar to the category of the query content from the database, and then is similar to the multidimensional space vector coordinate of the query content (the coordinate distance is the most Near) information having multidimensional space vector coordinates may be secondarily extracted from information extracted first.

In addition, the query and similarity analysis/extraction unit 360 performs a process of extracting a legal case related to the query content through the above-described process.

Thereafter, the average value of the positive and negative categories (numeric values) for each answer of at least one case precedent information and legal consultation information extracted from the query and similarity analysis/extraction unit 360 in the positive/negative quantification unit 370 If the positive or negative numerical value among the average values of the positive and negative categories (numeric values) of the calculation process and the extracted at least one case precedent information and legal consultation information is relatively more than a preset value, the numerical value reflecting the weight provides

Finally, the result value providing unit 380 provides a graph of the positive and negative numerical values calculated by the positive/negative digitizing unit 370 to the query terminal 200 in the form of a gauge. Here, the result providing unit 380 may use the numerical values of several cases as input values and provide them through a nonlinear model similar to the method of using stacking in final decision making in the ensemble technique. .

Therefore, by using the system and method for providing a numerical AI answer service to a legal question according to an embodiment of the present invention, the user can conveniently be curious about himself or have legal knowledge or legal knowledge about the events that have occurred to him. It has the advantage of being able to receive an answer on whether to apply it in the form of a gauge including positive and negative values, and to receive similar precedents and cases.

12 is a diagram illustrating an example computing environment in which one or more embodiments disclosed herein may be implemented, and is an illustration of a system 1000 including a computing device 1100 configured to implement one or more embodiments described above. shows For example, computing device 1100 may be a personal computer, server computer, handheld or laptop device, mobile device (mobile phone, PDA, media player, etc.), multiprocessor system, consumer electronics, minicomputer, mainframe computer, distributed computing environments including any of the aforementioned systems or devices, and the like.

The computing device 1100 may include at least one processing unit 1110 and a memory 1120 . Here, the processing unit 1110 may include, for example, a central processing unit (CPU), a graphic processing unit (GPU), a microprocessor, an Application Specific Integrated Circuit (ASIC), Field Programmable Gate Arrays (FPGA), etc. and may have a plurality of cores. The memory 1120 may be a volatile memory (eg, RAM, etc.), a non-volatile memory (eg, ROM, flash memory, etc.), or a combination thereof. Additionally, computing device 1100 may include additional storage 1130 . Storage 1130 includes, but is not limited to, magnetic storage, optical storage, and the like. The storage 1130 may store computer readable instructions for implementing one or more embodiments disclosed herein, and other computer readable instructions for implementing an operating system, an application program, and the like. Computer readable instructions stored in storage 1130 may be loaded into memory 1120 for execution by processing unit 1110 . Computing device 1100 may also include input device(s) 1140 and output device(s) 1150 .

Here, the input device(s) 1140 may include, for example, a keyboard, mouse, pen, voice input device, touch input device, infrared camera, video input device, or any other input device, or the like. Further, the output device(s) 1150 may include, for example, one or more displays, speakers, printers, or any other output device, or the like. Also, the computing device 1100 may use an input device or an output device included in another computing device as the input device(s) 1140 or the output device(s) 1150 .

Computing device 1100 may also include communication connection(s) 1160 that enable computing device 1100 to communicate with another device (eg, computing device 1300 ).

Here, communication connection(s) 1160 may be a modem, network interface card (NIC), integrated network interface, radio frequency transmitter/receiver, infrared port, USB connection, or other for connecting computing device 1100 to another computing device. It may include interfaces. Also, the communication connection(s) 1160 may include a wired connection or a wireless connection. Each component of the above-described computing device 1100 may be connected by various interconnections such as a bus (eg, peripheral component interconnection (PCI), USB, firmware (IEEE 1394), optical bus structure, etc.) and may be interconnected by a network 1200 . As used herein, terms such as "component," "system," and the like, generally refer to a computer-related entity that is hardware, a combination of hardware and software, software, or software in execution.

For example, a component can be, but is not limited to being, a process running on a processor, a processor, an object, an executable, a thread of execution, a program, and/or a computer. For example, both an application running on a controller and a controller may be a component. One or more components may reside within a process and/or thread of execution, and a component may be localized on one computer or distributed between two or more computers.

In the above, the present invention has been described in detail with reference to the embodiments, but those of ordinary skill in the art to which the present invention pertains may make various substitutions, additions and modifications within the scope not departing from the technical spirit described above. Of course, it should be understood that such modified embodiments also fall within the protection scope of the present invention as defined by the appended claims below.

100: A system that provides numerical AI answering services to legal questions
200: query terminal
300: AI legal / question and answer server
310: collection unit
320: preprocessor
330: database conversion unit
340: machine learning unit
350: natural language processing unit
360: Query and similarity analysis/extraction unit
370: positive / negative digitization unit
380: result value providing unit

Claims (20)

Inquiry terminal requesting legal advice on the content of the inquiry: and
After collecting case precedent information and legal consultation information of lawyers to which a case similar to the above query is applied from at least one external server, it is processed into a preset form and built into a database, and the established case precedent information and legal consultation information are stored After analyzing the legal answers to the affirmative and negative for the above query content using an artificial intelligence machine learning algorithm based on it, it includes an AI legal / question-and-answer server that provides numerical values of positive and negative values, respectively,
The system providing a numerical AI answering service to legal questions, characterized in that the query terminal executes a question and answer platform that displays the numerical value to the user in the form of a gauge.
According to claim 1,
The AI law/question answer server
After converting the case precedent information and legal consultation information collected from the at least one or more external servers into a question and answer format, the tendency of the answer is classified into either 'positive (positive)' or 'negative (passive)' A system that provides numerical AI answering services to legal questions built into a database.
3. The method of claim 2,
The AI law/question answer server
Numerical AI answering service to legal questions constructed so that at least one of raw data for the questions and answers, the category for the question, the case number, and the underlying law are included when building the questions and answers as a database system that provides
According to claim 1,
The AI law/question answer server
A system for providing a numerical AI answering service to a legal question, characterized in that it recognizes the tone of the user's query as either positive or negative.
4. The method of claim 3,
The AI law/question answer server
After calculating the sentence vectors of a plurality of questions and answers constructed in the database using the sentence vector similarity algorithm, the 10 sentences (questions and answers) with the highest similarity between sentences are selected, and the human After defining the perceived similarity as '0' or '1', learning the defined sentences using the Natural Language Processing Learning Algorithm (BERT NLP), re-learning the learned data to quantify the legal questions to derive learning data A system that provides an AI response service.
3. The method of claim 2,
The AI law/question answer server
A system for providing a numerical AI answer service to legal questions, characterized in that by using a natural language processing algorithm, semantic analysis of the query content, negative or positive tone, and a preset category are extracted.
7. The method of claim 6,
The AI law/question answer server
A system that provides a numerical AI answer service to a legal question that processes the sentence content (meaning) in the query content into multidimensional space vector coordinates using a natural language processing learning algorithm (BERT NLP).
8. The method of claim 7,
The AI law/question answer server
A system that provides a numerical AI answer service to a legal question that extracts case precedent information and legal consultation information including categories and multidimensional space vector coordinates similar to the category of the query content and multidimensional space vector coordinates from the database.
9. The method of claim 8,
The AI law/question answer server
A system for providing a numerical AI answer service to legal questions that provides the average value of the positive and negative categories (numeric values) of at least one extracted case precedent information and legal consultation information as the numerical value.
10. The method of claim 9,
The AI law/question answer server
A law providing a numerical value reflecting a weight when the positive or negative numerical value among the average values of the positive and negative categories (numeric values) of at least one or more extracted case precedent information and legal consultation information is relatively calculated to be greater than or equal to a preset value A system that provides numerical AI answering services to questions.
requesting legal advice (answer) on the contents of the inquiry from the inquiry terminal; and
In the AI law/question answer server, case precedent information and legal consultation information of lawyers to which cases similar to the above query are applied are collected from at least one external server, and then processed into a preset form to build a database, and the constructed case After analyzing the legal answers to the affirmation and negation of the query contents using an artificial intelligence machine learning algorithm based on precedent information and legal consultation information, providing a numerical value, wherein the query terminal provides the numerical value A method of providing a numerical AI answering service to legal questions, characterized in that it runs a question-and-answer platform that displays to the user in the form of a gauge.
12. The method of claim 11,
The step of providing the numerical value is
After converting the case precedent information and legal consultation information collected from the at least one or more external servers into a question and answer format, the tendency of the answer is classified into either 'positive (positive)' or 'negative (passive)' A method of providing a numerical AI answering service to legal questions including the step of building it into a database.
13. The method of claim 12,
The step of providing the numerical value is
Numericalization of legal questions, including the step of constructing to include at least one of raw data for the questions and answers, a category for the question, a case number, and the underlying law when building the questions and answers as a database How to provide a customized AI answer service.
12. The method of claim 11,
The step of providing the numerical value is
A method of providing a numerical AI answer service to a legal question, characterized in that it comprises the step of judging any one of positive and negative tone of the user's query content.
14. The method of claim 13,
The step of providing the numerical value is
After calculating the sentence vectors of a plurality of questions and answers constructed in the database using the sentence vector similarity algorithm, the 10 sentences (questions and answers) with the highest similarity between sentences are selected, and the human After defining the perceived similarity as '0' or '1', learning the sentences defined using a natural language processing learning algorithm (BERT NLP), and re-learning the learned data to derive learning data How to provide numerical AI answering service to questions.
13. The method of claim 12,
The step of providing the numerical value is
Using a natural language processing learning algorithm (BERT NLP) to analyze the meaning of the query, negative or positive tone, and extracting a preset category Provides a numerical AI answer service to legal questions How to.
17. The method of claim 16,
The step of providing the numerical value is
A method of providing a numerical AI answer service to a legal question, comprising processing the sentence content (meaning) in the query content into multidimensional space vector coordinates using the natural language processing learning algorithm (BERT NLP).
18. The method of claim 17,
The step of providing the numerical value is
A method of providing a numerical AI answer service to a legal question, comprising extracting from a database case precedent information and legal consultation information including categories and multidimensional space vector coordinates similar to the category of the query content and multidimensional space vector coordinates .
19. The method of claim 18,
The step of providing the numerical value is
A method of providing a numerical AI answer service to a legal question, comprising the step of providing an average value of positive and negative categories (numeric values) of at least one or more extracted case precedent information and legal consultation information as the numerical value.
20. The method of claim 19,
The step of providing the numerical value is
When the positive or negative numerical value among the average values of the positive and negative categories (numeric values) of the extracted at least one or more case precedent information and legal consultation information is relatively calculated to be greater than or equal to a preset value, providing a numerical value reflecting the weight How to provide a numerical AI answering service to legal questions that further includes.

Images