KR20210135831A - System for providing question and automatical answer service of homepage - Google Patents

Abstract

The present invention relates to a system for providing a service of automatically answering questions on a homepage. In accordance with an objective to be achieved, desired content among a large amount of content included in a homepage and a subdomain of the corresponding content are automatically extracted and provided as an answer to a question of a user. In accordance with one embodiment, the system includes: a question answer information service part extracting a characteristic value of a query inputted from a user, comparing the extracted characteristic value of the query to an answer keyword characteristic value of pre-learned answer content to select a URL of answer content which is most closely matched with the characteristic value of the query, and loading the selected URL to provide the corresponding content as an answer; and an answer content information learning part connecting the main URL of the homepage with a subdomain possessed by each of a plurality of pieces of answer content, and extracting a keyword included in the answer content of each subdomain and analyzing and learning a characteristic value thereof to provide the answer keyword characteristic value and a subdomain of the corresponding answer content to the question answer information service part.

Description

SYSTEM FOR PROVIDING QUESTION AND AUTOMATICAL ANSWER SERVICE OF HOMEPAGE

An embodiment of the present invention relates to a system for providing an automatic question answering service for a homepage.

In using the existing homepages of medical institutions such as hospitals or clinics, there is inconvenience in use as the user has to check all menus and information one by one to find the information that the user wants in the homepage.

In other words, the existing homepage provides several pages and shows detailed contents and information through each page, but in order to find the desired content, the user has to open all pages and click or select and read each content one by one. In the end, many users do not find the content they want and choose the method of calling the company on the website (for example, a medical institution such as a hospital or clinic) and consulting directly.

Accordingly, the operator of the website has to allocate a separate manpower for individual phone consultation and response, or if there is no such person, there are inconveniences and problems that interfere with the main business according to the phone consultation and response.

Registered Patent Publication No. 10-0775748 (Registration Date: November 05, 2007) Laid-Open Patent Publication No. 10-2009-0097304 (published date: September 16, 2009) Laid-Open Patent Publication No. 10-2016-0060253 (published date: May 30, 2016)

An embodiment of the present invention provides a system for providing an automatic answering service to a question on a homepage that automatically extracts a content containing a desired content and a sub-URL of the content from a lot of content included in the homepage to a user's query and provides it as an answer. to provide.

The system for providing an automatic query answering service for a homepage according to an embodiment of the present invention extracts a feature value of a query word input from a user, compares the extracted query word feature value with the answer keyword feature value of answer content learned in advance, and the query word a query answering information service unit that selects a URL of the answer content that most matches the feature value, calls the selected URL, and provides the corresponding content as an answer; and connecting the main URL of the homepage and the sub-URLs of the plurality of answer contents, extracting the keywords included in the answer contents of each sub-URL, analyzing and learning the characteristic values, and then analyzing and learning the characteristic values of the answer keywords in the question and answer information service unit. and an answer content information learning unit that provides a value and a sub-URL of the corresponding answer content.

In addition, the question and answer information service unit may include: a query input unit for receiving a query from a user; a query word feature analysis unit for extracting a keyword of the query word input through the query word input unit, synonyms and synonyms related to the corresponding keyword, respectively, and analyzing correlations between the extracted keywords, synonyms and synonyms to provide the query word feature values; and comparing the query word feature value provided through the query word feature analysis unit and the answer keyword feature value provided from the answer content information learning unit, selecting a sub-URL of the most matching answer content, and reading the selected sub-URL and an answer analysis providing unit that outputs the corresponding content as an answer to the query.

In addition, the query input unit may include at least one query input window so as to receive a query for each category according to an object receiving the query.

In addition, the answer content information learning unit connects the main URL of the homepage and the sub URLs of the plurality of answer content, respectively, registers a plurality of answer content on the homepage, and generates an answer learning database for generating an answer learning database with the registered answer content wealth; By reading the sub-URL of each answer content stored in the answer learning database, each of the keywords included in the answer content, synonyms and synonyms related to the keyword are extracted, and the correlation between the extracted keywords, synonyms and synonyms is analyzed. a response content keyword characteristic analysis unit providing the response keyword characteristic value; and machine-learning the answer keyword feature value provided by the answer content keyword feature analysis unit for the answer content stored in the answer learning database based on the correlation analyzed through the answer content keyword feature analysis unit, and include it in the learning result It may include an answer content machine learning unit for providing the sub-URL to the answer analysis providing unit.

In addition, the main URL of the homepage and the sub-URLs of the plurality of related contents are connected, and the characteristic value is analyzed by extracting related keywords included in the related contents of each sub-URL, and the analyzed related keyword characteristic value and the answer keyword A related content information providing unit may further include a related content information providing unit that matches between the characteristic values and provides a related keyword characteristic value matching the response keyword characteristic value and a sub-URL of the corresponding related content to the query and answer information service unit.

In addition, the event content information providing unit may include: a related content database generating unit configured to generate a related content database by registering related contents having sub URLs for each company operating a homepage; By reading the sub-URL of each related content stored in the related content database, each of the keywords included in the related content, synonyms and synonyms related to the keyword are extracted, and the correlation between the extracted keywords, synonyms and synonyms is analyzed. a related content keyword characteristic analysis unit providing the related keyword characteristic value; and matching between the relevant keyword feature value provided through the event content keyword feature analysis unit and the answer keyword feature value to provide a related keyword feature value matching the answer keyword feature value to the query answer information service unit, It may include a related content matching providing unit that provides a sub-URL of the related content having a keyword feature value.

In addition, the related content may include information about an event related to the answer content, precautions, finding a related company, benefits, related data, and related keywords.

According to the present invention, it is possible to provide a system for providing an automatic answering service to a question on a homepage that automatically extracts a content including a desired content and a sub-URL of the content from a lot of content included in the homepage to a user's query and provides it as an answer. can

1 is a schematic diagram illustrating an overall configuration relationship of a system for providing an automatic query answering service for a homepage according to an embodiment of the present invention.
2 is a block diagram showing the overall configuration of a system for providing an automatic query answering service for a homepage according to an embodiment of the present invention.
3 is a block diagram illustrating a detailed configuration of a question and answer information service unit according to an embodiment of the present invention.
4 is a block diagram illustrating a detailed configuration of an answer content information learning unit according to an embodiment of the present invention.
5 is a block diagram illustrating a detailed configuration of a related content information providing unit according to an embodiment of the present invention.
6 and 7 are diagrams illustrating an example of a method of inputting a query word in the query input unit and providing an answer by the answer content information learning unit according to an embodiment of the present invention.
8 to 11 are diagrams illustrating an example of an answer to a user's query and a screen for providing related content through the question and answer information service unit according to an embodiment of the present invention.
12 and 13 are diagrams illustrating an example of an answer providing screen to a user's query through a question and answer information service unit according to an embodiment of the present invention.

Terms used in this specification will be briefly described, and the present invention will be described in detail.

The terms used in the present invention have been selected as currently widely used general terms as possible while considering the functions in the present invention, but these may vary depending on the intention or precedent of a person skilled in the art, the emergence of new technology, and the like. In addition, in a specific case, there is a term arbitrarily selected by the applicant, and in this case, the meaning will be described in detail in the description of the corresponding invention. Therefore, the term used in the present invention should be defined based on the meaning of the term and the overall content of the present invention, rather than the name of a simple term.

When a part "includes" a certain element throughout the specification, this means that other elements may be further included, rather than excluding other elements, unless otherwise stated. In addition, terms such as "...unit" and "module" described in the specification mean a unit that processes at least one function or operation, which may be implemented as hardware or software, or a combination of hardware and software. .

Hereinafter, with reference to the accompanying drawings, embodiments of the present invention will be described in detail so that those of ordinary skill in the art can easily carry out the embodiments of the present invention. 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.

1 is a schematic diagram illustrating the overall configuration relationship of a system for providing an automatic query answering service for a homepage according to an embodiment of the present invention, and FIG. 1 is an overall system for providing an automatic question answering service for a homepage according to an embodiment of the present invention It is a block diagram showing the configuration, Fig. 3 is a block diagram showing the detailed configuration of the query and answer information service unit according to the embodiment of the present invention, Figure 4 is a block showing the detailed configuration of the answer content information learning unit according to the embodiment of the present invention 5 is a block diagram illustrating a detailed configuration of a related content information providing unit according to an embodiment of the present invention, and FIGS. 6 and 7 are a query input and answer content information learning unit of a query input unit according to an embodiment of the present invention. It is a view showing an example of an answer providing method, and FIGS. 8 to 11 are views showing an example of a screen for providing answers to a user's query and related contents through a question and answer information service unit according to an embodiment of the present invention, and FIG. 12 and FIG. 13 is a view showing an example of a screen for providing answers to a user's query through a question and answer information service unit according to an embodiment of the present invention.

1 and 2 , the system 1000 for providing an automatic question answering service for a homepage according to an embodiment of the present invention includes a question and answer information service unit 100 and an answer content information learning unit 200, and The related content information providing unit 300 may be further included.

The system 1000 for providing an automatic answering service to a question on the homepage of this embodiment is a homepage or a separate dedicated application that can be accessed through the main URL installed in the user terminal 10 and the homepage operator terminal 10, respectively, and for operating the overall system It can be done with a web server.

The user terminal 10 and the homepage operator terminal 10 are not only portable communication devices such as a smart phone, a smart pad, a tablet PC, etc., but also a general PC, and 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-Code Division Multiple Access (W-CDMA), and Wireless Broadband Internet (Wibro) terminals may be included.

The web server is a database-based web server, and has the same configuration as a conventional web server in terms of hardware, and is implemented in various types of languages such as C, C++, Java, Visual Basic, Visual C in terms of software. It can contain program modules that perform multiple functions. In addition, it can be implemented using web server programs that are provided in various ways according to operating systems such as DOS, Windows, Linux, Unix, and Macintosh on general server hardware. have.

In addition, the web server is connected to the user terminal 10 and the homepage operator terminal 10 through a network, and examples of such networks include RF, 3rd Generation Partnership Project (3GPP) network, Long Term Evolution (LTE) network, 5GPP. (5rd Generation Partnership Project) 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), a Bluetooth (Bluetooth) network, an NFC network, a satellite broadcasting network, an analog broadcasting network, a Digital Multimedia Broadcasting (DMB) network, etc. are included, but are not limited thereto.

The query and answer information service unit 100 extracts a feature value of a query word input from the user, compares the extracted query word feature value with the answer keyword feature value of the answer content learned in advance, and answers the most consistent with the query word feature value You can select the URL of the content, call the selected URL, and provide the corresponding content as an answer.

To this end, the question and answer information service unit 100 may include a query input unit 110 , a query word characteristic analysis unit 120 , and an answer analysis providing unit 130 as shown in FIG. 3 .

The query input unit 110 may be configured to receive a query input from a user through the input window of the homepage 20 , At least one query input window may be provided to receive a query for each category according to an object to be inquired.

For example, as shown in FIG. 6 , the query input unit 110 may receive a query word in a word or natural language form from the user through the input window, and may specify a subject for receiving a question, that is, a doctor or a nurse. , guide, parking, plastic surgery, orthopedics, counseling, gun construction, exercise, etc. If you can specify the person being questioned, you can select one of the many icons representing the person receiving the question and then enter the query. As shown in FIG. 7 , on the basis of the content database of categories divided by icons, it is possible to provide each professional answer from each category icon, so that it is possible to provide faster and more accurate answers. Even if the user does not select such a category icon, the category of the answer area to the corresponding question may be automatically classified through the analysis of the query word.

On the other hand, when the homepage 20 is executed through a mobile device like the user terminal 10 , the query input unit 110 may input the query word by voice even if it is not input through the keyboard through the food recognition function of the mobile device.

The query word characteristic analysis unit 120 may extract the keyword of the query word input through the query word input unit 110, and then extract synonyms and synonyms related to the query word keyword from a pre-built database, respectively, the extracted keyword, A query word feature value may be provided by analyzing a correlation between synonyms and/or synonyms.

For example, if a query word "What is bimaxillary surgery?" is inputted, the keyword 'two jaw surgery' is extracted from each of the query words, and among them, a synonym or synonym for 'chin' related to the query word 'positive medicine'. can be extracted from the database, and by analyzing the correlation between 'Western surgery' and 'Chin surgery', we extract or create feature values for query words such as 'Double Jaw Surgery Definition', 'Two Jaw Surgery Method', 'Western Surgery Hospital', etc. can do.

The answer analysis providing unit 130 compares the query word feature value provided through the query word feature analyzer 120 with the answer keyword feature value provided from the answer content information learning unit 200, and matches the query word feature value the most. After finding the answer keyword feature value, select the sub URL of the answer content with the answer keyword feature value based on the machine learning algorithm, read the selected sub URL and convert the content of the sub URL into the homepage ( 20) can be printed.

For example, as shown in FIG. 8 , when a query word “Tell me the type of LASIK surgery” is input through the query word input unit 110, a sub URL having an answer keyword feature value that most matches the keyword feature value of the query word It is possible to read and output response contents such as '3D Smile, LASIK & LASIK' through the corresponding URL through the homepage 20 .

On the other hand, the answer analysis providing unit 130 may provide the content of the understanding of the user's query together with the answer content. For example, as shown in FIGS. 12 and 13 , between the query word content and the answer content, in order to explain that the user understands the content of the query, a brief comment stating that the user is curious about what content is and provides an answer to the content can be generated and displayed.

The answer content information learning unit 200 connects the main URL of the home page 20 and the sub-URLs of the plurality of answer content, extracts keywords included in the answer content of each sub-URL, and analyzes the feature value and By learning, it is possible to provide an answer keyword feature value and a sub-URL to the question and answer information service unit 100 .

To this end, the answer content information learning unit 200 may include an answer learning database generator 210 , an answer content keyword characteristic analysis unit 220 , and an answer content machine learning unit 230 as shown in FIG. 4 . .

The answer learning database generation unit 210 registers a plurality of answer contents in the homepage 20 by connecting the main URL of the homepage 20 and the sub URLs of the plurality of answer contents, respectively, and learns the answer with the registered answer contents. A database 211 may be created.

More specifically, if there is a homepage 20 for operating a specific hospital or clinic, the homepage 20 is classified as a medical institution and is the URL of the homepage 20 so as not to be confused with homepages for other purposes. The main URL may be linked to each other so as to have a relationship with the sub-URL, which is the URL of the answer contents related thereto. That is, according to the category and operating purpose of the operating company, group, or institution of the homepage 20 , the sub-URL may be registered and connected to the main URL so that the answer content related thereto is utilized as data or information of the corresponding homepage.

The answer content keyword characteristic analysis unit 220 reads sub-URLs of all answer content stored in the answer learning database 211 and extracts keywords, corresponding keyword-related synonyms and/or synonyms, respectively, included in the answer content, An answer keyword feature value may be provided by analyzing the correlation between the extracted keywords, synonyms, and synonyms.

For example, if the answer content contains information related to orthognathic surgery, keywords used in the content and synonyms and/or synonyms related to the keyword are extracted, and the relationship between them is analyzed to determine the It is possible to calculate and provide the characteristic values of answer keywords such as 'Surgery definition', 'Western medicine surgery method', and 'Two jaw surgery simulation'.

The answer content machine learning unit 230, the answer content keyword characteristic analysis unit 220 for the answer content stored in the answer learning database 211 based on the correlation analyzed through the answer content keyword characteristic analysis unit 220 It is possible to machine-learning the answer keyword feature value provided in , and provide a sub-URL of the answer content included in the learning result to the answer analysis providing unit 130 .

The related content information providing unit 300 connects the main URL of the homepage 20 and the sub-URLs of the plurality of related contents, extracts related keywords included in the related contents of each sub-URL, and analyzes the characteristic value. Then, by matching between the analyzed related keyword feature value and the answer keyword feature value, it is possible to provide the related keyword feature value matching the answer keyword feature value and the sub-URL of the related content to the query and answer information service unit.

To this end, the related content information providing unit 300 may include a related contents database generating unit 310 , a related contents keyword characteristic analysis unit 320 , and a related contents matching providing unit 330 as shown in FIG. 5 . .

The related content database generating unit 310 may generate a related content database 311 by registering related contents having sub URLs for each company operating the homepage 20 .

More specifically, if there is a homepage 20 for operating a specific hospital or clinic, the homepage 20 is classified as a medical institution and is the URL of the homepage 20 so as not to be confused with homepages for other purposes. The main URL may be linked to each other so as to have a relationship with the sub-URL, which is the URL of related contents. That is, according to the category and operating purpose of the operating company, group, or organization of the homepage 20 , the related content related thereto is information about the event, precautions, related company search, benefits, related data, and related keywords of the homepage 20 . It can be connected by registering a sub URL in the main URL so that it can be used as a .

The related content keyword characteristic analysis unit 320 reads a sub-URL of each related content stored in the related content database 311 and extracts keywords included in the related content, synonyms and synonyms related to the keyword, respectively, and extracts A related keyword feature value may be provided by analyzing the correlation between the used keywords, synonyms, and synonyms.

The related content matching providing unit 330 matches between the related keyword feature value provided through the event content keyword feature analyzing unit 330 and the answer keyword feature value to provide the answer keyword feature value to the query and answer information service unit 100 . It is possible to provide a related keyword feature value that matches with a sub-URL of a related content having a corresponding related keyword feature value.

This related content information providing unit 300, as well as delivering information by providing answer content through each homepage 20, events, notices, related In order to provide information on company search, benefits, related data, and related keywords, keywords included in the answer content or additional information related to the content may be provided. After matching the related content related to the content, when the user's desired answer information is provided, the related information may be additionally provided together.

For example, as shown in FIG. 8 , when the answer content 30 for the query 'Tell me the type of LASIK surgery' is provided, at the same time, events related to the answer content (LASIK surgery) 30, precautions, It is possible to provide related content information such as hospital search, insurance benefits, and surgical method, and each related content is provided in the same form as a keyword tag 41. Through this, you can proceed with related events or check related information additionally.

In addition, as shown in FIG. 9 , the related content 40 may be provided as an event or guide information related thereto after checking the answer content for 'LASIK surgery type', and each content (event hospital guide, medical insurance coverage) It is provided in the form of a menu that can be selected for each information) so that the user can obtain the desired event or additional information.

In addition, as shown in FIG. 10, items such as 'Register my child', 'View vaccination history', and 'Find a designated medical institution' are provided as related content 40 for the query 'I wonder when my child will be vaccinated'. Since there is a sub-URL for each item, when a user selects (clicks or touches) a desired item, the user can move to a page where the procedure for the item can be performed or create a new page.

On the other hand, as shown in FIG. 11 , when a query word 'Tell me about a hospital that does an event' is input, the query is related to the related content of the system 1000, so as an answer to this, the sub-URL of the related content is found and various hospitals It may be provided for access to the website of the House of Representatives and the House of Representatives.

What has been described above is only one embodiment for implementing the system for providing an automatic question answering service for a homepage according to the present invention, and the present invention is not limited to the above embodiment, but as claimed in the claims below Without departing from the gist of the invention, it will be said that the technical spirit of the present invention exists to the extent that various modifications can be made by anyone with ordinary knowledge in the field to which the invention pertains.

1000: A system for providing an automatic answering service to a question on the homepage
100: Q&A Information Services Department
110: query input unit
120: query language feature analysis unit
130: answer analysis providing unit
200: answer content information learning unit
210: answer learning database generator
211: Answer Learning Database
220: answer content keyword characteristic analysis unit
230: Answer content machine learning unit
300: related content information providing unit
310: related content database generation unit
311: Associated content database
320: related content keyword characteristic analysis unit
330: related content matching providing unit
10: user terminal / homepage operator terminal
20: Homepage
30: Answer content
40: Related content
41: Related keywords

Claims (7)

Extracting the feature value of the query word input by the user, comparing the extracted query word feature value with the answer keyword feature value of the pre-learned answer content, selecting the URL of the answer content that most matches the query word feature value, and selecting the selected URL a question-and-answer information service unit that calls and provides the corresponding content as an answer; and
The main URL of the homepage and the sub-URLs of each of the plurality of answer contents are connected, the keywords included in the answer contents of each sub-URL are extracted, the characteristic value is analyzed and learned, and the response keyword characteristic value is analyzed in the question and answer information service unit. and an answer content information learning unit that provides a sub URL of the corresponding answer content.
According to claim 1,
The question and answer information service unit,
a query input unit receiving a query input from a user;
a query word feature analysis unit for extracting a keyword of the query word input through the query word input unit, synonyms and synonyms related to the corresponding keyword, respectively, and analyzing correlations between the extracted keywords, synonyms and synonyms to provide the query word feature values; and
Compare the query word feature value provided through the query word feature analysis unit and the answer keyword feature value provided from the answer content information learning unit, select a sub-URL of the most matching answer content, and read the selected sub-URL An automatic question answering service providing system for the homepage, characterized in that it includes an answer analysis providing unit that outputs the corresponding content as an answer to the query.
3. The method of claim 2,
The query input unit,
A system for providing an automatic query answering service for a homepage, comprising at least one query input window to receive a query by category according to an object to be inquired.
3. The method of claim 2,
The answer content information learning unit,
an answer learning database generating unit for registering a plurality of answer content on the homepage by connecting the main URL of the homepage and sub URLs of the plurality of answer content, respectively, and generating an answer learning database from the registered answer content;
By reading the sub-URL of each answer content stored in the answer learning database, each of the keywords included in the answer content, synonyms and synonyms related to the keyword are extracted, and the correlation between the extracted keywords, synonyms and synonyms is analyzed. a response content keyword characteristic analysis unit providing the response keyword characteristic value; and
Based on the correlation analyzed through the answer content keyword feature analysis unit, machine learning the answer keyword feature value provided by the answer content keyword feature analyzer for the answer content stored in the answer learning database is machine-learned, and included in the learning result and an answer content machine learning unit for providing a sub-URL to the answer analysis providing unit.
5. The method of claim 4,
The main URL of the homepage and the sub-URLs of the plurality of related contents are connected, and the characteristic value is analyzed by extracting the related keywords included in the related contents of each sub-URL, and the analyzed related keyword characteristic value and the answer keyword characteristic value Automatic answer to a query on the homepage, characterized in that it further comprises a related content information providing unit for matching between the two and providing the related keyword characteristic value matching the response keyword characteristic value and a sub-URL of the corresponding related content to the query answering information service unit service delivery system.
6. The method of claim 5,
The event content information providing unit,
a related content database generating unit generating a related content database by registering related contents having sub URLs for each company operating a homepage;
By reading the sub-URL of each related content stored in the related content database, each of the keywords included in the related content, synonyms and synonyms related to the keyword are extracted, and the correlation between the extracted keywords, synonyms and synonyms is analyzed. a related content keyword characteristic analysis unit providing the related keyword characteristic value; and
Matching between the relevant keyword feature value provided through the event content keyword feature analysis unit and the answer keyword feature value to provide a relevant keyword feature value matching the answer keyword feature value to the query answer information service unit, the relevant keyword A system for providing an automatic query answering service for a home page, comprising a related content matching providing unit that provides a sub-URL of a related content having a characteristic value.
6. The method of claim 5,
The related content is
The system for automatically answering questions on the homepage, characterized in that it includes information on events related to the answer content, precautions, search for related companies, benefits, related materials, and related keywords.

Images