KR102086156B1 - User customized system and method for searching and matching expert - Google Patents

Abstract

According to an aspect of the present invention, there is provided a user-customized expert matching and retrieval system and method thereof, comprising: a user terminal configured to receive predetermined information by a user and enable wired / wireless data networking; And a platform for storing content information or expert information, and a network connection with the user terminal to transmit the content information or the expert information to the user terminal when requested by the user terminal. In addition to the search, the platform analyzes the posts uploaded by the user and provides user-specific information according to the keywords of the posts.

Description

User customized system and method for searching and matching expert}

The present invention relates to a system and a method for allowing a user to search and match an expert in a customized manner. More specifically, after extracting a keyword through registration information uploaded by a user, a legal expert or the like suitable for the user's registration information is provided. The present invention relates to a system and method for providing a user's expert information or content information in accordance with registration information uploaded by a user or the like.

As society and industry are advanced and diversified, various experts in each field are active at various levels.

Among the legal services sectors, the number of legal service professionals in 2007 was 3548 attorney businesses, which was 4871 in 2014. The number increased from 24,552 to 3,555,3 in 2014 (see "Science and Technology Services Survey Data," Statistics Korea).

Although the quantitative growth of legal experts is supported by the government's policy to provide high quality and low-cost legal services to the majority of people, access to these experts is still difficult and inconvenient for the general public.

This is because even legal experts are divided into various fields to provide professional services in different areas of law, and it is difficult for the general public to easily know their detailed areas of expertise proposed by these legal experts. .

Various technical attempts have been made to connect general law consumers and legal experts, such as "a system and method for providing expert knowledge-based search service (Registration No. 10-1596307, hereafter 'Patent Document 1')". .

Patent document 1 collects knowledge from experts in each field through video calls, reprocesses the collected video call data, extracts keywords, and then builds a content DB using the extracted keywords. The present invention relates to a system and a method for providing a search service based on expert knowledge.

Another technical attempt is "expert service matching system (publication No. 10-2014-0145239, hereinafter referred to as" Patent Document 2 ").

Patent document 2 is an expert service matching method that can implement a market environment that satisfies the needs of consumers through the Internet, the process of registering a professional handling field of the expert; Expert selection process for selecting M experts in the field of specialized handling requested by the consumer; An expert extraction process of randomly extracting N experts from the selected M experts; Notifying the extracted N experts to request a response to the consultation contents; Matching system when both the expert's area and professional handling area and the consumer's area and accident area are met simultaneously, including the process of granting the consumer the right to respond only to the first expert who is notified of the response. As a result, consumers living in regions other than the Seoul metropolitan area can effectively provide expert assistance.

Also related technologies include "expert search system and method (Publication No. 10-2014-0000408, hereinafter referred to as 'Patent Document 3').

In case of Patent Literature 3, an expert pool generation module for analyzing a network user's activity in the network, ranking experts selected by specialty according to the analysis result, and constructing an expert pool, and searching an expert pool according to a user's query And an expert search module for providing the user with a list of experts derived from the search results, and a feedback module for updating the expert pool using the expert evaluation results input from the user.

However, these prior arts provide information about whether the consumer needs information is simply a search of a legal expert, a match with a legal expert, information about cases or precedents similar to his situation, As a result of the lack of analysis and the ability to provide services for form documents, most of the inconveniences of general law consumers still exist.

Registration No. 10-1596307 Publication No. 10-2014-0145239 Publication No. 10-2014-0000408

The user-customized expert matching and retrieval system and method thereof according to the present invention have been devised to solve the conventional problems as described above, and present the following problems to be solved.

First, it basically presents a search function of legal experts, but provides a customized service according to the analyzed user posts by analyzing the posts uploaded by the user.

Second, we will provide customized information by analyzing the posts uploaded by the user.

Third, we will analyze the user's various information over time and provide user customized information.

The problems of the present invention are not limited to those mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the following description.

The user-customized expert matching and retrieval system and method thereof according to the present invention have the following problem solving means for the above-mentioned problem.

According to an aspect of the present invention, there is provided a user-customized expert matching and retrieval system comprising: a user terminal receiving predetermined information from a user, and having a wired / wireless data networking function; And a platform that stores content information or expert information, and transmits the content information or the expert information to the user terminal when the user terminal requests a network connection with the user terminal.

The platform of the user-customized expert matching and retrieval system according to the present invention may be characterized by outsourcing the content information or the expert information from an external server.

The platform of the user-customized expert matching and retrieval system according to the present invention provides a search function of the content information or the expert information to the user terminal to transfer the content information or the expert information requested by the user terminal to the user terminal. It may include a search unit to provide.

The platform of the user-customized expert matching and retrieval system according to the present invention includes an information registration unit for receiving and storing registration information uploaded by the user terminal; An information extraction unit for extracting a plurality of keywords from the registration information; And an information presenting unit which provides the content information or the expert information associated with the plurality of keywords to the user terminal based on the plurality of keywords extracted by the information extracting unit.

The information extraction unit of the user-customized expert matching and retrieval system according to the present invention may be characterized by assigning weights according to the degree of frequentness to the extracted plurality of keywords.

The information extracting unit of the user-customized expert matching and retrieval system according to the present invention may be configured to reweight the plurality of keywords to which the weight is assigned based on a preset term.

According to an aspect of the present invention, there is provided a user-customized expert matching and retrieval system comprising: an information analyzer configured to accumulate the registration information and produce history information based on a time course of the registration information; And an information linker for providing the expert information and the content information according to the history information to the user terminal.

The information linker of the user-customized expert matching and retrieval system according to the present invention provides the history information to an external server, receives prediction information according to the history information from the external server, and provides the prediction information to the user terminal. It can be characterized by.

The prediction information of the user-customized expert matching and retrieval system according to the present invention may be calculated through a statistical estimate of the history information.

The statistical estimate of the user-customized expert matching and retrieval system according to the present invention may be set to information of an area mapped onto a plurality of data tables linked to the history information.

According to an aspect of the present invention, there is provided a user-customized expert matching and searching method comprising: inputting predetermined information to a user terminal by a user (a1); (A2) receiving, by a platform networking with the user terminal, the predetermined information; And (a3), by the platform storing content information or expert information, transmitting the content information or the expert information corresponding to the predetermined information to the user terminal.

The platform of the user-customized expert matching and retrieval method according to the present invention may be characterized by outsourcing the content information or the expert information from an external server.

The predetermined information of the user-customized expert matching and retrieval method according to the present invention may be a search word that may be mapped to the expert information or the content information.

According to an aspect of the present invention, there is provided a user-customized expert matching and searching method comprising: inputting registration information to a user terminal by a user (b1); (B2) receiving, by a platform networking with the user terminal, the registration information; Extracting a plurality of keywords from the registration information by the platform (b3); And based on the extracted plurality of keywords, the platform may include providing content information or expert information related to the keyword to the user terminal (b4).

The user-customized expert matching and retrieval method according to the present invention may further include, after step (b3), assigning a weight according to the degree of frequent occurrence appearing in the extracted plurality of keywords (b3-1). Can be.

The user-customized expert matching and retrieval method according to the present invention further includes, after step (b3-1), re-weighting the plurality of extracted keywords based on preset terminologies (b3-2). It can be characterized.

According to an aspect of the present invention, there is provided a user-customized expert matching and retrieval method comprising: accumulating and storing the registration information and producing history information according to a time course of the registration information; And providing the expert information or the content information according to the history information to the user terminal (b6).

The user-customized expert matching and searching method according to the present invention includes the step (b5-1) of providing the linkage information to an external server after the step (b5); Receiving prediction information following the linkage information from the external server (b5-2); The method may further include transmitting the prediction information to the user terminal (b5-3).

The prediction information of the user-customized expert matching and retrieval method according to the present invention may be calculated through a statistical estimate of the history information.

The statistical estimate of the user-customized expert matching and retrieval method according to the present invention may be set to information of an area mapped onto a plurality of data tables linked to the history information.

The user-customized expert matching and searching system and method thereof according to the present invention having the above-described configuration provide the following effects.

First, the user can directly search for legal experts, the platform analyzes the posts uploaded by the user, and provides user-specific information according to the keywords of the posts.

Second, the analysis of the user's posts, for example, providing the legal expert to the user according to the analysis of the keywords, providing the case case, such as case and case information, or providing a document form file suitable for the user's situation, etc. To provide.

Third, through predicting the legal service information needed in the near future through the statistical technique through the user's profile information and the history of the article posted by the user, the user's personalized information is presented preemptively accordingly.

The effects of the present invention are not limited to those mentioned above, and other effects that are not mentioned will be clearly understood by those skilled in the art from the following description.

1 is a conceptual diagram illustrating a relationship between a user terminal, expert information, and content information, a platform providing a user-customized expert matching and retrieval system according to the present invention, and a method thereof.
FIG. 2 is a block diagram illustrating an operation relationship of components of a user-customized expert searching and matching system according to an embodiment of the present invention.
3 is a block diagram showing a database corresponding to one component of the present invention and a lower DB (database) constituting the same.
4 is a conceptual diagram illustrating an operation relationship between a user terminal, a central server, and a database according to an embodiment of the present invention.
5 is a conceptual diagram illustrating an operation relationship between a user terminal, a central server, and a database according to another embodiment of the present invention.

The user-customized expert matching and retrieval system and method thereof according to the present invention can be variously modified and have various embodiments. Specific embodiments will be illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the present invention to specific embodiments, it should be understood to include all changes, equivalents, and substitutes included in the technical spirit and scope of the present invention.

1 is a conceptual diagram illustrating a relationship between a user terminal, expert information, and content information, a platform providing a user-customized expert matching and retrieval system according to the present invention, and a method thereof. FIG. 2 is a block diagram illustrating a functional relationship between a user-customized expert and components of a matching system according to an embodiment of the present invention. 3 is a block diagram showing a database corresponding to one component of the present invention and a lower DB (database) constituting the same. 4 is a conceptual diagram illustrating an operation relationship between a user terminal, a central server, and a database according to an embodiment of the present invention. 5 is a conceptual diagram illustrating an operation relationship between a user terminal, a central server, and a database according to another embodiment of the present invention.

According to the present invention, the user-customized expert search and matching system, as shown in FIG. 1, allows a user to search for an expert, for example, a legal service expert, and content information and expert information that can be generally provided in the legal service area. It has a form of the platform 100 for providing a user to fit the postings and the like.

As shown in FIG. 1, the user-customized expert search and matching system according to the present invention includes a user terminal 10 and a platform 100, where the user terminal 10 is a general user, that is, a user. As a terminal used by a user, the user may be a general PC, a laptop computer, a tablet personal computer, a smart phone, or the like.

The user terminal 10 is a terminal having a function of wired / wireless data networking, and is provided to enable wired / wireless data networking with the platform 100 mentioned in the present invention.

The user terminal 10 may be provided with various information input means to receive predetermined information input by the user and upload it to the platform 100.

The various information input through the user terminal 10 may be a search word for searching for expert information or content information, or may be in the form of a post uploaded by a user.

As shown in FIGS. 2 and 3, the platform 100 may have a database 102 of itself, and may store various pieces of information such as content information 30 or expert information 20. In addition, the platform 100 may outsource these content information 30 or expert information 20 through an external server (not shown).

Here, the content information 30 may be a document form file for various legal reports or applications, as well as legal information, precedent information, legal case information, and legal interpretation information when the specialized area to be handled is a legal service area. .

Here, the expert information 20 refers to a comprehensive information on the lawyer and related information such as the attorney's education, histories, the legal area handled professionally, the prevailing rate, the number, etc., when the professional area to be handled is the legal service area. .

The external server refers to a server that allows the platform 100 to communicate data with an external server through the central server 101 so that the platform 100 can query and pull necessary information. Search portal sites like Naver or Google can be any server that handles specialized information.

The platform 100 previously stores the content information 30 or the expert information 20 as described above, or outsources it from an external server, and requests the user terminal 10 to the user terminal 10 when the user terminal 10 requests it. Provide the requested information.

As shown in FIG. 2, the user-customized expert search and matching system according to the present invention may include a searcher 110.

When the user terminal 10 inputs a search word, the search unit 110 provides a search function to the user terminal 10 to search the content information 30 or expert information 20 searched by the user terminal 10. It is provided to the user terminal 10.

Search unit 110 is not only a configuration that provides a search function mainly focused on search terms and search operators that can be provided, but also a sentence input by the user using a natural language processing API (Application Programming Interface) provided by Google, etc. It can also recognize and provide a search function.

As shown in FIG. 2, the user-customized expert searching and matching system according to the present invention includes an information register 120; Information extracting unit 130; And an information presenting unit 140.

First, in the case of the information registration unit 120, the user provides a function of receiving and storing registration information, which is information such as a post written and uploaded by the user terminal 10. The user terminal 120 provides the user with an interface for registering information, for example, an interface for writing a post, and through this function, the user writes and uploads specialized information, for example, legal matters, to be inquired by the user. These posts will be uploaded on the platform 100 by the user terminal 10 through the information register 120 of the platform 100.

The information extraction unit 130 recognizes the registration information received and stored by the information registration unit 120 and extracts a plurality of keywords included in the registration information. After recognizing nouns, adjectives, investigations, adverbs, and verbs, the extracted keywords are reinterpreted from complex sentences to formalized patterned sentences through formalization and patterning functions according to subjects, objects, and verbs. Through this, keywords are extracted based on nouns that are repeatedly spoken.

Subsequently, the information presenting unit 140 provides the content information 30 or the expert information 20 related to these keywords based on the plurality of keywords extracted by the information extracting unit 130 to the user.

For example, if the registration information input by the user, that is, a keyword extracted by the information extracting unit 130 in a post is mainly composed of "accident," "vehicle," "traffic," "consensus," "error rate," etc. The presenting unit 140 may provide content information 30 related to a traffic accident, for example, traffic accident case information, various application document form files related to a traffic accident, and expert information 20 related to a traffic accident. You will be provided with comprehensive information on attorneys, etc.

As another example, a user-entered registration information, that is, a keyword extracted by the information extraction unit 130 from a post is "spouse," "affair," "confidence," "divorce," "alimony," "trial divorce," " In the case of a divorce agreement, "etc., the information presenting unit 140 may provide content information 30 or expert information 20 related to divorce.

In more detail, the information extraction unit 130 may increase the accuracy of the information provided by the information presentation unit 140 by assigning a weight to the extracted plurality of keywords according to the degree of frequent occurrence.

For example, among the registered information input by the user, keywords such as "Affair", "Affair", etc. are higher than the frequency of the keyword for "Violence" among keywords such as "Infidelity," "Affair," "Violence," etc. In this case, a weight is assigned to “affair” or “infidelity”, and in this case, the information presenter 140 provides case information related to affair and affair rather than providing precedent information including domestic violence and divorce information. This will increase the accuracy of the information to be provided.

Furthermore, the information extracting unit 130 may re-weight the plurality of weighted keywords based on preset terminology. In the case of the pre-determined terminology here, the legal terminology that the user knows and enters in advance, and if the weight is actively added to the terminology, the accuracy of the customized information to be provided can be increased. Because there is.

For example, if a legal term "contract divorce" is included in the registration information entered by the user, the need for consultation about the divorce of consultation is stronger than the general divorce method. As such, the terminology may be reweighted to increase the accuracy of the information to be provided.

The terminology may be a terminology provided by the platform 100 itself or a term that is outsourced through an external server, or a term previously input by an expert linked to a service provided by the platform 100.

As shown in FIG. 2, the user-customized expert search and matching system according to the present invention includes an information analyzer 150; And an information linker 160.

In the case of the information analyzer 150, the registration information uploaded by the user through the user terminal 10 is accumulated, and the history information is processed according to the time course of the registration information.

First, the first registration information entered by the user mainly includes keywords for "aint," "trial divorce," "consensus divorce," "partition of property," "an alimony." In the case of the predominantly divorce, "custody," "parental rights," etc., the information analysis unit (in addition to the content information 30 about "trial divorce" provided by the information presenting unit 140 in the second registration information) 150) is a historical information, and in addition to considering various cases, the current user may or may not proceed with a trial divorce process, and may provide various legal information, that is, content information, during the trial divorce process that may be requested in the future. By making predictions, these predictions are used as one piece of historical information.

In this case, the information linking unit 160 provides the predicted information as linkage information such as the plaintiff or the defendant in the trial divorce proceedings, and the legal information, precedent information, and the like after the divorce proceedings. In addition, various case information that may be generated during the property division process may be additionally provided before the user requests.

The information linker 160 may provide such history information to an external server, receive the above-described prediction information, and provide the same to the user terminal 10. In this process, the operator side of the external server may be provided. And non-disclosure agreements (NDAs) to prevent users from leaking or infringing their personal information.

Here, the prediction information may allow the platform 100 mentioned in the present invention to have its own artificial machine learning algorithm, and thus new registration information following existing registration information according to accumulation of history information. The feedback may be designed to increase the accuracy of the prediction information.

The external server described above may of course enable the user to analyze big data recorded on other external servers through the Spark of Apache. Of course, big data collected and stored by an external server can be collected through Hadoop's Greenplum.

The prediction information provided by the information linker 160 according to the history information may be calculated through the above-described statistical estimation through machine learning, and the statistical estimation may be mapped onto a plurality of data tables linked to the history information. It may be set to the information of the area to be.

In this case, the data table (not shown) may be divided first by category of the legal service area, and in the second category, legal information according to time is divided. First, in the case of the legal service area divided into the first, it may be a construction lawsuit, traffic lawsuit, divorce lawsuit, medical lawsuit, corporate M & A, bankruptcy and rehabilitation procedures, and in the case of the second division, the first divorce lawsuit is " , "" Contract divorce ", etc., and can be divided into three types, such as" procedures, "" cases, "and" property consultation split "within the consultation divorce. Through the area of these three-dimensional data tables, the history information calculates predictive information that can be subsequently followed according to the registered information of the user uploaded at a specific point of time through statistical estimates, thereby improving the accuracy of providing customized information according to the user. You can increase it.

As shown in FIG. 4, in the user-customized expert searching and matching method according to the present invention, the platform 100 for data networking with the user terminal 10 as described above may receive predetermined information such as a search word. To have.

Thereafter, in more detail than the platform 100, the central server 101 of the platform 100 makes a request for information according to the search word to the database 102, and then the database requests these requests via the central server 101. The expert information 20 or the content information 30 according to the information is provided to the user terminal 10.

In the case of FIGS. 2 and 4, the expert information 20 and the content information 30 are shown as being sourcing only through the database 102 disposed inside the platform 100, but the platform 100 shows the content information. Of course, the 30 or expert information 20 can be outsourced through an external server.

As shown in FIG. 4, it is assumed that the above-mentioned predetermined information is a search word that may correspond to the expert information 30 or the content information 20.

Furthermore, according to the present invention, a user-specific expert search and matching method, as shown in Figure 5, after the registration information is input to the user terminal 10 by the user, networking with the user terminal 10 The platform 100 may have a step of receiving registration information.

As described above, the registration information here is upload information in the form of a post input by a user.

Thereafter, the platform 100 undergoes a step of extracting a plurality of keywords from the registered information, and then, based on the extracted plurality of keywords, the platform 100 may include content information 30 or expert information 20 associated with the keyword. ) Is provided to the user terminal 10.

Subsequently, the user-customized expert searching and matching method according to the present invention may further include a step of assigning a weight according to the degree of frequent occurrence in the extracted plurality of keywords.

Subsequently, the user-customized expert searching and matching method according to the present invention may further include re-weighting the extracted plurality of keywords based on preset terminology.

The user-specific expert search and matching method according to the present invention accumulates and stores registration information uploaded by the user through the user terminal 10, and generates history information according to the time course of registration information and history information. The method may further include providing the expert information 30 or the content information 20 to the user terminal 10.

Subsequently, the user-customized expert searching and matching method according to the present invention may include providing linkage information to an external server, receiving prediction information following linkage information from the external server, and providing the prediction information to the user terminal 10. The method may further include transmitting to.

The prediction information may be calculated through statistical estimates of the above-described history information, and the statistical estimate may be set through information of a region mapped onto a plurality of data tables linked to the history information.

The scope of the present invention is determined by the matters set forth in the claims, and the parentheses used in the claims are not for the purpose of selective limitation, but rather for the sake of clarity. Should be.

10: user terminal 20: expert information
30: content information 100: platform
101: Central Server 102: Database
110: search unit 120: information register
130: information extraction unit 140: information presentation unit
150: information analysis unit 160: information linking unit

Claims (20)

A user terminal receiving predetermined information from a user and having a wired / wireless data networking function; And
A platform for storing content information or expert information and transmitting the content information or the expert information to the user terminal when the user terminal requests a network connection with the user terminal;
The platform,
An information registration unit for receiving and storing registration information uploaded by the user terminal;
An information extraction unit for extracting a plurality of keywords from the registration information; And
An information presenting unit providing the content information or the expert information associated with the plurality of keywords to the user terminal based on the plurality of keywords extracted by the information extracting unit;
An information analyzer configured to accumulate the registration information and produce history information according to a time course of the registration information; And
And an information linker which provides the expert information and the content information according to the history information to the user terminal.
The information extraction unit,
Weights according to the degree of frequentness are first given to the extracted plurality of keywords,
The information extraction unit,
Among the plurality of keywords to which the weight is primarily assigned, the weight is secondarily re-adjusted for the same keyword as a preset term, and is corrected.
The information link unit,
Providing the history information to an external server, receiving prediction information according to the history information from the external server, and providing the prediction information to the user terminal,
The prediction information,
Calculated through statistical estimates of the history information,
The history information,
Primary category information including primary keywords extracted from the registration information primarily input by a user; And
After the extraction of the primary keyword, extracted from the registration information input by the user secondary and includes secondary category information including a secondary keyword having a sub-concept for the primary keyword,
The prediction information,
Including tertiary category information including a tertiary keyword predicted from the primary category information and the secondary category information through machine learning and having a sub-concept for the secondary keyword,
The information presenting unit is a user-specific expert search, characterized in that to present to the user the expert information and the content information corresponding to the third category information, together with the expert information and the content information corresponding to the secondary category information. And matching system.
The method of claim 1, wherein the platform,
Outsourcing the content information or the expert information from an external server.
The method of claim 1, wherein the platform,
And a search unit for providing the content information or the expert information to the user terminal to provide the content information or the expert information requested by the user terminal to the user terminal. Matching system.
delete delete delete delete delete delete The method of claim 1, wherein the statistical estimate is:
And a user-customized expert search and matching system, characterized in that it is set as information of an area mapped onto a plurality of data tables linked to the history information.
delete delete delete (B1) inputting registration information to the user terminal by the user;
(B2) receiving, by a platform networking with the user terminal, the registration information;
Extracting a plurality of keywords from the registration information by the platform (b3);
Based on the extracted plurality of keywords, the platform providing content information or expert information associated with the keyword to the user terminal (b4);
Accumulating and storing the registration information and producing history information according to a time course of the registration information; And
(B6) providing the expert information or the content information according to the history information to the user terminal,
After the step (b3),
(B3-1) firstly assigning weights according to the degree of frequentness to the extracted plurality of keywords; And
(B3-2) further re-weighting and correcting weights of the same keywords as the preset terminology among the plurality of keywords to which the weights are primarily applied;
After the step (b5),
Providing the history information to an external server (b5-1);
Receiving prediction information according to the history information from the external server (b5-2);
Transmitting the prediction information to the user terminal (b5-3),
The prediction information,
Calculated through statistical estimates of the history information,
The history information,
Primary category information including primary keywords extracted from the registration information primarily input by a user; And
After the extraction of the primary keyword, extracted from the registration information input by the user secondary and includes secondary category information including a secondary keyword having a sub-concept for the primary keyword,
The prediction information,
Including tertiary category information including a tertiary keyword predicted from the primary category information and the secondary category information through machine learning and having a sub-concept for the secondary keyword,
Step (b4) is,
The user customized expert search and matching is performed by presenting the expert information and the content information corresponding to the third category information to the user together with the expert information and the content information corresponding to the second category information. Way.
delete delete delete delete delete The method of claim 14, wherein the statistical estimate is:
The user-customized expert search and matching method, characterized in that the information set in the area (mapping) mapped on the plurality of data tables linked to the history information.

Images