KR102107303B1 - System and method for providing product manual based on chatbot - Google Patents

Abstract

The present invention provides a chatbot-based interactive interface to a user, and provides a chatbot service providing module that provides an answer to a user's product related inquiries through the interactive interface; A conversation history DB (database) for storing conversation content made in the interactive interface; Manual DB that stores product manuals of multiple companies; And when product information and product related inquiries are input from the user terminal through the interactive interface, search for answers to the product related inquiries in the conversation history DB with reference to the product information, and search for the searched answer contents. A search module provided by the chatbot service providing module; It relates to a chatbot-based product manual providing system comprising a.

Description

System and method for providing product manual based on chatbot {SYSTEM AND METHOD FOR PROVIDING PRODUCT MANUAL BASED ON CHATBOT}

The present invention relates to a system and method for providing a product manual based on a chatbot, and more particularly, to a system and method for providing a product manual based on a chatbot capable of conveniently providing product manual information and the like through a chatbot service.

The product manual (manual) is a user's manual for the product, and performs the functions of the product's guidance, explanation, and instructions on the product, such as how to use the product, the main functions of the product, precautions, and how to take action in case of failure. These product manuals were usually provided in booklet form when the product was purchased.

However, the manual in the form of a booklet is difficult to solve immediately for product problems because portability is poor and it is difficult to quickly find the content desired by the user.

In addition, the user may access the manufacturer's homepage or search the Internet separately, but in this case, it takes a lot of time and effort to solve the product problem. Often the manual could not be found by the method.

The present invention is to provide a chatbot-based product manual providing system and method for conveniently obtaining information and a manual of a product desired by a user by providing a chatbot-based interactive service to a user.

In addition, another object of the present invention is to provide a system and method for providing a chatbot-based product manual that can formally store and manage multiple product manuals.

In addition, another object of the present invention is to provide a system and method for providing a chatbot-based product manual that can manage the product manuals of small and medium-sized enterprises in an integrated manner, thereby enhancing the convenience of use and enhancing competitiveness of SME products.

The chatbot-based product manual providing system according to an embodiment of the present invention provides a chatbot-based interactive interface to a user, and a chatbot service providing module that provides an answer to a user's product-related inquiry through the interactive interface When product information and product related inquiries are input from the user terminal through the interactive interface DB (database) for storing conversation contents made in the interactive interface, a manual DB storing product manuals of multiple companies, and the interactive interface, The search module may include a search module that searches for an answer to the product related inquiry in the conversation history DB with reference to the product information and provides the searched answer to the chatbot service providing module.

In addition, if the search module does not search for an answer to the product-related inquiry in the conversation band DB or if there is an additional search request from the user terminal, the search module displays a specific product manual corresponding to the product information in the manual DB. Search, and search for the answers to the product related inquiries from the searched specific product manuals to provide the chatbot service providing module.

In addition, the chatbot-based product manual providing system further includes a question-and-answer DB for storing question-and-answer information collected from the manufacturer's homepage of the product, and the search module identifies a specific information corresponding to the product information in the manual DB. If the product manual is not searched or there is an additional search request from the user terminal, the answer information for the product related inquiry is searched in the query and answer DB with reference to the product information, and the searched answer information is provided to the chatbot service module Can be provided as

In addition, if the search module does not search for an answer to the product related inquiry in the query response DB or there is an additional search request from the user terminal, the search module answers the product related inquiry by referring to the product information. The content can be searched online and the searched response content can be provided to the chatbot service providing module.

Further, the product information may include a manufacturer of the product, a name of the product, and a model name of the product.

In addition, the chatbot service providing module may request information about at least one of a manufacturer of the product, a name of the product, and a model name of the product to the user through the chatbot.

In addition, the chatbot-based product manual providing system, the related keyword DB where product-related keywords are stored for each product, and the search module use the product information input through the interactive interface to search for product-related keywords related to the product. Search in the related keyword DB, and the searched product related keyword may be provided to the chatbot service providing module.

In addition, the chatbot service providing module provides a product related keyword provided from the search module to the user through the chatbot, and recognizes a product related keyword selected by the user among the product related keywords as the product related inquiry It can be sent to the search module.

In addition, it may further include a manual collection unit for collecting product manuals or online product manuals provided by the company and storing them in the manual DB.

In addition, further comprising a data extraction unit for extracting the structured data from the product manual stored in the manual DB, the structured data, product-related keywords, the manufacturer of the product, the name of the product, the model name of the product, and the manufacturer It may include the homepage address.

In addition, using the address of the manufacturer's homepage extracted by the data extractor, access to the manufacturer's homepage, and collect a question and answer information existing in the manufacturer's homepage and store it in the query and answer DB further It can contain.

The method for providing a chatbot-based product manual according to an embodiment of the present invention provides an interactive interface providing a chatbot-based interactive interface to a user, and product information and product related inquiries from the user terminal through the interactive interface. If it is input, includes a step of searching for an answer to search for an answer to the product-related inquiry with reference to the product information, and a step of providing an answer for providing the user with the searched answer through the interactive interface. In the answer content search step, an answer content for the product-related inquiry may be searched in a conversation history DB (database) that stores the conversation content made in the interactive interface.

In addition, in the step of searching for the answer contents, if the answer contents for the product related inquiries are not searched in the conversation band DB or an additional search request is requested from the user terminal, the product is stored in the manual DB storing product manuals of multiple companies. A specific product manual corresponding to information may be searched for, and an answer content for the product related inquiry may be searched for in the searched specific product manual.

In addition, in the answer content search step, when a specific product manual corresponding to the product information is not searched in the manual DB or an additional search request is requested from the user terminal, the product information is referred to for the product related inquiries. The contents of the answer can be searched in the question and answer DB, and the question and answer DB can store question and answer information collected from the manufacturer's homepage of the product.

In addition, in the step of searching for the answer contents, if the answer contents for the product related inquiries are not searched in the query response DB or there is an additional search request from the user terminal, refer to the product information to answer the product related inquiries. You can search for answers online.

Further, the product information may include a manufacturer of the product, a name of the product, and a model name of the product.

In addition, in the step of providing the interactive interface, the user may request information about at least one of a manufacturer of the product, a name of the product, and a model name of the product.

In addition, using the product information input through the interactive interface further comprises a related keyword search step of searching for a product related keyword related to the corresponding product in the related keyword DB, wherein the related keyword DB includes a product related keyword It can be stored very much.

In addition, in the providing of the interactive interface, a product related keyword searched in the related keyword search step may be provided to a user, and a product related keyword selected by the user among the product related keywords may be recognized as the product related inquiry. .

In addition, a manual collection step of collecting a product manual provided by a company or an online product manual and storing the product manual in the manual DB may be further included.

In addition, the structured data extraction step of extracting the structured data from the product manual stored in the manual DB further includes, the structured data, product-related keywords, the manufacturer of the product, the name of the product, the model name of the product, and the It may include the manufacturer's homepage address.

In addition, a query characterized by accessing the manufacturer's homepage by using the manufacturer's homepage address extracted in the structured data extraction step, and collecting and answering the question and answer information existing in the manufacturer's homepage in the query response DB It may further include a response collection step.

According to the present invention, by providing a chatbot-based interactive service to a user, it is possible to provide a chatbot-based product manual providing system and method for conveniently obtaining information and a manual of a product desired by a user.

In addition, according to the present invention, it is possible to provide a chatbot-based product manual providing system and method that can formally store and manage multiple product manuals.

In addition, according to the present invention, it is possible to integrally manage product manuals of SMEs, thereby providing a chatbot-based product manual providing system and method for enhancing the convenience of using and enhancing the competitiveness of SME products.

1 is a view showing a chatbot-based product manual providing system according to an embodiment of the present invention.
2 is a view showing a detailed configuration of a chatbot-based product manual providing system according to an embodiment of the present invention.
3 is a view for explaining a chatbot-based interactive interface according to an embodiment of the present invention.
4 is a view showing a part of the product manual by way of example.
5 exemplarily shows question and answer information according to an embodiment of the present invention.
6 is a diagram illustrating a product-related keyword according to an embodiment of the present invention.
7 is a view for explaining a chatbot-based interactive interface according to another embodiment of the present invention.
8 is a view for explaining the extraction of product-related keywords from the product manual according to an embodiment of the present invention.
9 is a view for explaining extracting product-related keywords from a product manual according to another embodiment of the present invention.
10 is a view for explaining the extraction of the manufacturer of the product, the name of the product, the model name of the product, and the like from the product manual according to an embodiment of the present invention.
11 is a diagram for explaining extraction of a product model name or the like according to another embodiment of the present invention from a product manual.
12 is a flowchart illustrating a method for providing a chatbot-based product manual according to an embodiment of the present invention.
13 is a flowchart illustrating an answer content search step according to an embodiment of the present invention.

Hereinafter, exemplary embodiments related to the present invention will be illustrated in the drawings and described in detail through detailed description. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various different forms, and should be understood as including all modifications, equivalents, and substitutes included in the spirit and scope of the present invention. .

In describing the components of the present invention, terms such as first, second, A, B, (a), and (b) may be used. These terms are only for distinguishing the component from other components, and the nature, order, or order of the component is not limited by the term. Further, in the present specification, when a component is described as being “connected”, “coupled” or “connected” to another component, the component may be directly connected to or connected to the other component, but each component It will be understood that another component may be "connected", "coupled" or "connected" between the elements. In the case of "connection", "coupled" or "connection", it is understood that physically "connected", "coupled" or "connected", as well as electrically "connected", "coupled" or "connected" as needed Can be.

The terms "~ unit (unit)", "~ group", "~ character", and "~ module" described in the present specification mean a unit that processes at least one function or operation, which is hardware or software or hardware. And software. In addition, terms such as “comprise”, “consist” or “have” described in this specification mean that the corresponding component can be inherent, unless specifically stated to the contrary, excluding other components. It should not be interpreted as being able to further include other components.

In addition, it is intended to clarify that the division of the constituent parts in this specification is only classified according to main functions of each constituent part. That is, two or more components to be described below may be combined into one component, or one component may be divided into two or more for each subdivided function. In addition, each of the components to be described below may additionally perform some or all of the functions of other components in addition to the main functions of the components, and some of the main functions of the components are different. Needless to say, it may also be carried out exclusively by.

Hereinafter, a system for providing a chatbot-based product manual and a method according to an embodiment of the present invention will be described with reference to drawings related to embodiments of the present invention.

1 is a view showing a chatbot-based product manual providing system according to an embodiment of the present invention.

Referring to FIG. 1, the chatbot-based product manual providing system 100 according to an embodiment of the present invention is connected to a network, thereby transmitting / receiving information to or from various terminals 200 or through various information (eg, product manuals) ).

Here, the network means a connection structure capable of exchanging information between each node, such as a terminal and a server, and an example of such a network includes the Internet, a local area network (LAN), and a wireless local area (LAN). Network (WAN), Wide Area Network (WAN), Personal Area Network (PAN), 3G, 4G, 5G, LTE, Wi-Fi, and the like. The Internet is a TCP / IP protocol and various services that exist in the upper layer, such as HyperText Transfer Protocol (HTTP), Telnet, File Transfer Protocol (FTP), Domain Name System (DNS), Simple Mail Transfer Protocol (SMTP), and SNMP ( It means an open computer network structure that provides Simple Network Management Protocol (NFS), Network File Service (NFS), and Network Information Service (NIS).

The terminal 200 may access a chatbot-based product manual providing system 100 according to an embodiment of the present invention through a network, and a company, product user, and system administrator through the chatbot-based product manual providing system 100 ) Can use and manage the services provided.

For example, the terminal 200 may include an enterprise terminal 210, a user terminal 220, and an administrator terminal 230.

In the case of the corporate terminal 210, a terminal used by a company that manufactures and sells products, and a corporate representative provides a product manual management request system based on a chatbot based product manual and a product manual management system through the corporate terminal 210 (100).

The user terminal 220 is a terminal used by a user who purchases a product, and the product user can access the chatbot-based product manual providing system 100 through the user terminal 220, through which the user manual of the product is systemized. Search and query on the (100), or the information they currently want can be provided from the system (100).

The administrator terminal 230 is a terminal used by the administrator of the chatbot-based product manual providing system 100, and the system administrator performs a management function of the corresponding system 100 through the administrator terminal 230 or when a specific situation occurs Notification messages and the like may be collected from 100.

The terminal 200 is not particularly limited as long as it is a device having a communication function connected to the chatbot-based product manual providing system 100 and a display function capable of outputting an image or text. For example, the terminal 200 is a desktop computer, a laptop computer, a tablet PC, a wireless phone, a mobile phone, or a smart phone. , Smart watch, smart glass, e-book reader, portable multimedia player (PMP), portable game machine, navigation device, digital camera, digital multimedia broadcasting (DMB) player , Digital audio recorder, digital audio player, digital picture recorder, digital picture player, digital video recorder, digital video player ( digital video player), but is not limited thereto.

The chatbot-based product manual providing system 100 may collect product manuals online through a network. For example, the chatbot-based product manual providing system 100 may periodically or continuously search the product manual in the form of a file online, and collect the searched product manual.

In addition, the chatbot-based product manual providing system 100 may collect product manuals in a file format provided by a company. For example, a company that wants to register a product manual to a chatbot-based product manual providing system 100 may upload a corresponding product manual to the system 100 through the corporate terminal 210.

The product manual providing system 100 based on the chatbot can collect product manuals of a number of companies in the same manner as described above, and can perform standardization to extract and construct structured data for the product manuals.

In addition, the chatbot-based product manual providing system 100 may provide a chatbot-based interactive interface to the user terminal 220, through which a product-related inquiry may be input from the user and an answer may be provided. .

2 is a view showing a detailed configuration of a chatbot-based product manual providing system according to an embodiment of the present invention.

Referring to FIG. 2, the chatbot-based product manual providing system 100 according to an embodiment of the present invention may include a chatbot service providing module 110, a search module 120, and a database 130. .

The chatbot service providing module 110 may provide a chatbot-based interactive interface between the user terminal 220 and the system 100, and for this purpose, the user terminal 220 may provide a chat window that the user and chatbot can communicate with. Can be. For example, the chatbot-based interactive interface may be made through mobile messenger, web messenger, message (SMS), and the like.

The chatbot service has made great leaps forward with big data analysis, machine learning, and natural language processing technologies to infer the correct answer based on the analysis of the user's conversation content. The following questions are predicted to provide a conversation service, and more recently, due to the convenience of use, the service has expanded beyond a simple conversation service to use for various operations such as shopping and payment.

The chatbot service providing module 110 may provide a chatbot-based interactive interface to a product user who has difficulty in using the product, and the user enters a product-related inquiry to be solved and the chatbot of the interactive interface responds to this. By answering, you can conveniently solve problems related to the use of the product.

In addition, the chatbot service providing module 110 may store the conversation history between the chatbot and the user in the conversation history DB 131 through the corresponding interface, and refer to the conversation scenario stored in the conversation scenario DB 136 to communicate between the chatbot and the user. You can proceed with the conversation. The chatbot service providing module 110 may be provided in the form of an application programming interface (API).

When the product information and product-related inquiries are input from the user terminal 220 through the chatbot-based interactive interface provided by the chatbot service providing module 110, the search module 120 searches for an answer to this and searches again It can be provided by the chatbot service providing module 110. The chatbot service providing module 110 provided with such an answer content may deliver the corresponding answer content to the user through the chatbot.

The database 130 may store collected manuals, various structured data extracted from the manuals, and various information for answering product related inquiries.

For example, the database 130 includes a conversation history DB 131, a manual DB 132, a question and answer DB 133, an associated keyword DB 134, a product information DB 135, a conversation scenario DB 136, etc. It may include.

The conversation history DB 131 may store conversation history between a chatbot and a user in a chatbot-based interactive interface provided by the chatbot service providing module 110.

The manual DB 132 may store product manuals of multiple companies, and the Q & A DB 133 may store Q & A information collected from a manufacturer's homepage.

The associated keyword DB 134 may store product-related keywords for each product, and the product information DB 135 may store various information related to products. For example, the product information DB 135 may store a product manufacturer, a product name, a product model name, a manufacturer's homepage address, and a manufacturer's phone number.

The conversation scenario DB 136 may store a conversation scenario of the chatbot existing in the interactive interface. With reference to this conversation scenario, the corresponding chatbot proceeds with the conversation, and through this, information for solving a product problem can be requested or transmitted to a user.

The manual collection unit 140 may collect product manuals provided by a company or online product manuals.

For example, the manual collection unit 140 may receive the product manual from the enterprise terminal 210 or collect the product manual through a method of searching the product manual online.

In addition, the manual collection unit 140 may store the collected product manuals in the manual DB 132.

The data extraction unit 150 extracts structured data from the product manuals collected by the manual collection unit 140 to perform a standardization operation on the product manuals.

At this time, the standardized data extracted from the product manual as a result of standardization may include product-related keywords, the manufacturer of the product, the name of the product, the model name of the product, the address of the manufacturer's homepage, and additionally the manufacturer's phone number. It can contain.

In addition, the data extraction unit 150 stores the extracted product-related keywords in the associated keyword DB 134, the product-related information of the product manufacturer, product name, product model name, manufacturer's homepage address, manufacturer's phone number Etc. can be stored in the product information DB 135.

The question and answer collection unit 160 accesses the manufacturer's homepage using the manufacturer's homepage address extracted by the data extraction unit 150, collects the question and answer information existing in the manufacturer's homepage, and answers the question and answer DB (133) ).

In general, the Q & A information is a 'Frequently Asked Questions (FAQ)' in which the manufacturer or the seller combines the questions and answers, and / or a 'user who provides the answers to the manufacturer or the seller when the user fills out an inquiry 'Q & A'.

Therefore, the Q & A collection unit 160 may associate the question and answer existing in the corresponding homepage and store the Q & A in the Q & A DB 133 together.

3 is a view for explaining a chatbot-based interactive interface according to an embodiment of the present invention, Figure 4 is a view showing a part of the product manual by way of example, Figure 5 is a question and answer according to an embodiment of the present invention This is a diagram showing the question and answer information stored in the DB by way of example.

The user may access the chatbot-based interactive interface provided by the chatbot service providing module 110 using the user terminal 220.

For example, the user may launch a messenger application installed on the user terminal 220, search for 'manual chatbot', and start a conversation with the chatbot.

The user can enter an inquiry related to a product he / she currently desires in a chat window provided by the interactive interface.

For example, referring to FIG. 3, a user who wants to know how to change the door lock password can input 'Change door lock password' in the corresponding chat window.

In this case, the chatbot service providing module 110 may request the product information necessary for specifying the product to the user through the chatbot. The request of the chatbot may be made of an option as shown in FIG. 3 or only a request sentence without an option.

For example, since the user has already mentioned the name of the product 'door lock', the chatbot service providing module 110 may sequentially request the user of the manufacturer of the product and the model name of the product.

Usually, the specification of a product may be made of product information of a manufacturer of a product, a name of a product, and a model name of a product, and different dialogue scenarios may be performed according to a user's input order. The progress of the conversation may be performed with reference to the conversation scenario stored in the conversation scenario DB 136 described above.

For example, if the user inputs only product information without inquiries, after specifying the product, the user can be inquired about the product-related inquiries.

In addition, when product information is not input, a query for product information may be performed, and when specific information is first input among product information, a query may be performed for the remaining product information as illustrated in FIG. 3.

Such a dialogue scenario may be variously changed, and it is sufficient to receive predetermined product information and product related inquiries for specifying a product through an interactive interface for implementation of the present invention.

When the input of the manufacturer of the product and the model name of the product is completed by the user, the search module 120 searches the answer history for the product-related inquiry in the conversation history DB 131 by referring to the product information provided by the user , The searched response content may be transmitted to the chatbot service providing module 110.

For example, the search module 120 displays the conversation history DB 131 whether there is a conversation history about the password change method (product related inquiries) related to TR610 (model name) among the door locks (names of products) of Hyegang Security (manufacturer). You can search.

If there is an answer to the password change method for the corresponding product, the search module 120 may transmit the corresponding answer to the chatbot service providing module 110 so that it can be provided to the user through the chatbot.

However, if the corresponding answer is not searched or the user wants additional search, the search module 120 searches for a specific product manual corresponding to the product information in the manual DB 132, and inquires about the product in the searched specific product manual You can search the contents of the answer.

Referring to FIG. 4, the search module 120 can search the product manual for the door lock TR610 model of Hyegang Security from the manual DB 132 and search for the contents of the password change method described on a specific page of the product manual. have.

At this time, the search module 120 extracts only text information from the searched part in the product manual and provides it to the chatbot service providing module 110 or extracts the searched part as an image and provides it to the chatbot service providing module 110. .

The chatbot service providing module 110 that receives the text and / or image corresponding to the response content may transmit the corresponding response content to the user by causing the chatbot to speak the corresponding response content.

In addition, the search module 120 may refer to the product information by referring to the product information when a specific product manual corresponding to product information input from a user in the manual DB 132 is not searched or an additional search request is requested from the user terminal 220. The contents of the answers to the related inquiries can be searched in the Q & A DB 133.

For example, referring to FIG. 5, the Q & A DB 133 may store Q & A information related to the door lock collected from the manufacturer's (Hyegang Security) homepage. At this time, since the question and answer information may be classified by manufacturer and / or product, the search module 120 may perform a search by referring to at least a part of the product information.

Accordingly, the search module 120 can find the query response information of the manufacturer (Hyegang Security) in the Q & A DB 133, and extracts the answer related to the password change from the question and answer information to provide the chatbot service module 110 Can be provided as

At this time, the question and answer information may be composed of text and / or images, and the chatbot service providing module 110 that receives the text and / or images corresponding to the answer contents allows the chatbot to speak the corresponding answer contents, thereby responding to the corresponding answers. Content can be delivered to the user.

As described above, the search module 120 may sequentially search the conversation history DB 131, the manual DB 132, and the question and answer DB 133 to derive the answers to product-related inquiries.

Nevertheless, if there is no answer content for the product-related inquiry in the Q & A DB 133 or the user does not want the answer, the search module 120 Can search online for answers to product-related questions.

At this time, the search module 120 may search for the answer content using a portal service such as Naver or a search service such as Google.

At this time, the content of the searched answer may include text and / or images, and the chatbot service providing module 110 that receives the text and / or image corresponding to the content of the answer allows the chatbot to speak the content of the response, so that the response content Can be delivered to the user.

On the other hand, the chatbot service providing module 110 can provide 'detailed view', 'complete manual view', and 'relevant video view' functions when providing answers to product related inquiries to the user through the chatbot's remarks. have.

For example, when the user selects (clicks or selects) the 'detailed view' link displayed in the chat window, the chatbot service providing module 110 displays a page including a detailed description of the corresponding answer and / or related images. It can be provided to the user terminal 220.

In addition, when the user selects (click or selects) 'View the entire manual' displayed on the chat window, the chatbot service providing module 110 provides the manual of the corresponding product to the user terminal 220 so that the user can view the entire manual. I can do it.

In addition, when the user selects (click or selects) 'view related video' displayed in the chat window, the chatbot service providing module 110 provides a video related to the answer to the user terminal 220 so that the user can view the related video. You can make it watchable.

6 is a diagram illustrating a product-related keyword according to an embodiment of the present invention by way of example, and FIG. 7 is a view for explaining a chatbot-based interactive interface according to another embodiment of the present invention.

Referring to FIG. 6, the associated keyword DB 134 may store product-related keywords for information related to products or inquiries that the user wants to solve for each product.

For example, for 'door lock' products, 'password registration and change method', 'smart key registration and change method', 'smart key deletion method', 'open door from outside', 'door lock from outside', 'indoor' In 'Open door', 'Lock door in the room', 'Forced lock function', etc. can be designated as product related keywords.

These product-related keywords may be extracted from the product manual collected by the data extraction unit 150 as described above, and a detailed extraction method of the related keywords will be described later.

Referring to FIG. 7, the chatbot service providing module 110 may provide a product related keyword for determining a user's interest item.

To this end, when the user inputs product information (for example, the name of a product), the search module 120 searches the product related keywords related to the corresponding product in the related keyword DB 134, and searches the product related keywords that are searched. It can be provided by the chatbot service providing module 110.

For example, a user who has started a conversation with a chatbot by accessing the interactive interface can enter the name of the product he owns, 'Door Lock', through the corresponding chat window.

In response to this, the search module 120 may search for a product related keyword related to the door lock in the related keyword DB 134 and provide it to the chatbot service providing module 110.

Accordingly, the chatbot service providing module 110 is' 1. How to register and change password ',' 2. How to register and change smart key ',' 3. How to delete smart key ',' 4. Open outdoors ',' 5. Lock the door outdoors ',' 6. Open indoors ',' 7. Lock the door indoors, '8. Force lock function ', etc. can be provided in the corresponding chat window as a keyword related to the product.

For example, if the user wants to know how to delete the smart key of the door lock, the user can select a keyword related to the product. The selection of product related keywords can be made by entering a number or a corresponding product related keyword in a chat window.

If the user selects 'Door lock smart key deletion method', the chatbot service providing module 110 may recognize the product related keyword as a product related inquiry and transmit it to the search module 120.

In response to this, the search module 120 displays the contents of the response to the inquiry (the smart key deletion method of the door lock) as described above, the conversation history DB 131, the manual DB 132, and the question and answer DB 133, It can be derived by searching online information.

In FIG. 7, a dialogue in which other product information (such as a product manufacturer and a product model name) is input is omitted in addition to the product name, but in this case, a chatbot for user identification of the product and a user-to-user dialogue can also be made by referring to the dialogue scenario. .

8 is a view for explaining the extraction of product-related keywords from the product manual according to an embodiment of the present invention.

Referring to FIG. 8, the data extraction unit 150 may extract product related keywords 310 and 320 from the collected product manuals.

For example, the data extraction unit 150 may extract product-related keywords 310 and 320 from a table of contents of a product manual stored in the manual DB 132.

In general, a table of contents for content guidance exists in product manuals, and important keywords representing each page or each item of a product manual exist in these pages of contents.

Accordingly, the data extraction unit 150 may select keywords described in the table of contents as product-related keywords 310 and 320, and the selected product-related keywords 310 and 320 may be stored in the storage 130. Can be saved in association with the manual.

For example, 'product specifications and components', 'name of each part', 'usage method', and 'convenient function (optional function)' constituting the table of contents page may be extracted as the keyword 310 related to the first product. In addition, 'Password registration and change method', 'Smart key registration and change method', 'Smart key deletion method', 'Open outdoors' , 'Locking the door outdoors', 'Opening the door indoors', 'Locking the door indoors', 'Forced lock function', 'Set and release the lock status check function', etc., will be extracted as the keywords related to the second product (320) Can be.

Also, the 'automatic / manual lock setting', which further subdivides the 'convenient function (selection function)' that is the first product related keyword 310, may also be extracted as the second product related keyword 320.

When the operation of extracting product-related keywords 310 and 320 of the data extraction unit 150 is described in more detail, the data extraction unit 150 may first search the table of contents page in the product manual.

For example, the data extraction unit 150 may select a page including a table of contents estimation keyword as a table of contents. For example, the keyword for estimating the table of contents may include 'table of contents', 'table of contents', 'contents', and the like. However, the present invention is not limited thereto, and words indicating a table of contents may be additionally included.

After the table of contents page is selected in the corresponding product manual, the data extraction unit 150 extracts a string composed of the largest character among the strings excluding the table of contents estimation keyword from the table of contents page as the first product related keyword 310, and Next, a character string composed of the second largest character may be extracted as the second product related keyword 320.

9 is a view for explaining extracting product-related keywords from a product manual according to another embodiment of the present invention.

Depending on the product manual, there may be a case where there is no table of contents page. In this case, since the keyword of the table of contents is not searched in the product manual, the data extraction unit 150 may determine that there is no table of contents page in the corresponding product manual.

When it is determined that there is no table of contents page in the corresponding product manual, the data extraction unit 150 may extract product-related keywords 310 and 320 from each page of the product manual.

For example, referring to FIG. 9, a string composed of the largest character in each page of the product manual is the first product-related keyword 310 and a character string composed of the second-largest character in the second product-related keyword ( 320).

This is because, in general, in the product manual, the title or phrase representing the page is displayed as the largest character, and in the case of its sub-title or phrase, it is often displayed as a slightly smaller character.

Through the above method, the data extraction unit 150 may extract product related keywords 310 and 320 even when a table of contents page does not exist in the product manual.

10 is a view for explaining the extraction of the product manufacturer, product name, product model name, etc. from the product manual according to an embodiment of the present invention, and FIG. 11 is a model name of a product according to another embodiment of the present invention It is a diagram for explaining extracting from a product manual.

The data extraction unit 150 may extract structured data such as a product manufacturer 410, a product name 420, and a product model name 430 from a product manual. In addition, the data extracting unit 150 may store the structured data such as the manufacturer 410 of the extracted product, the name 420 of the product, and the model name 430 of the product in association with the corresponding manual in the storage unit 130. .

For example, the data extracting unit 150 may search for a character string including a manufacturer's estimated keyword in the product manual, and extract the character string to the manufacturer 410 of the product. As an example, the manufacturer's estimated keyword may include 'stock company', '(shares)', and the like. However, the present invention is not limited thereto, and words indicating the manufacturer may be additionally included.

Accordingly, as shown in FIG. 10, 'Daum Soft Co., Ltd.' including the manufacturer's estimated keyword '(Note)' may be extracted to the manufacturer 410 of the product.

In addition, the data extraction unit 150 may extract the name 420 of the product from the product manual by referring to the list of products held.

For example, the data extraction unit 150 may select the name 420 of the product through comparison of words and product lists in the product manual.

For example, if the product list includes 'digital door lock', the data extracting unit 150 can recognize that the phrase 'digital door lock' searched in the product manual is the product name, and accordingly the corresponding product name (420 ).

If the word matching the product list does not exist in the product manual, the data extraction unit 150 may transmit a notification message informing the administrator terminal 230.

Accordingly, the system administrator can determine the name of the target product by reviewing the corresponding product manual. Thereafter, the system administrator may transmit the product name determined through the manager terminal 230 to the chatbot-based product manual providing system 100, and the data extraction unit 150 may transmit the received product name to the product name of the corresponding product manual ( 420).

In addition, the data extraction unit 150 may update the product name list by adding the product name received from the manager terminal 230 to the product name list.

Meanwhile, the data extracting unit 150 divides the range of the product manual into the product manufacturer 410 and the product name 420 to speed up the extraction speed of the product manufacturer 410 and the product name 420. Search can be performed sequentially.

For example, the data extracting unit 150 may perform a search first for the first and last pages of the product manual, where the product manufacturer 410 and the product name 420 are likely to exist. After the search, if it is determined that the manufacturer 410 of the product and / or the name 420 of the product does not exist, then a search for the remaining pages may be performed.

Meanwhile, the model name 430 of the product is generally composed of a combination of English letters and numbers. Accordingly, the data extraction unit 150 may extract a character string including English letters and numbers from the product manual as the model name 430 of the product.

However, the data extraction unit 150 according to an embodiment of the present invention may search for the model name 430 of the product by adding additional conditions to increase the accuracy of extraction.

For example, as shown in FIG. 10, the model name 430 of the product mainly exists on the first page of the product manual, so the data extraction unit 150 is a character string containing alphabets and numbers for the first page of the product manual. Search to extract the product's model name (430).

Accordingly, in FIG. 10, 'DMT-100' may be extracted as the model name 430 of the product.

In addition, since the model name 430 of the product is mainly located adjacent to the name 420 of the product, the data extraction unit 150 is located adjacent to the name 420 of the extracted product. It can be extracted by the model name 430 of.

Accordingly, in FIG. 11, 'CHA-DA10' located adjacent to the 'electric insulating rice cooker' extracted as the product name 420 may be extracted as the model name 430 of the product.

When the number of character strings extracted according to the model name extraction operation according to the above-described method is large, it can be considered that some noise data is included. Accordingly, the character string including the preset negative keyword 431 may be excluded, and the model name 430 of the product may be determined from the remaining character strings.

For example, except for the string including the negative keyword 431, the remaining string can be extracted directly into the product's model name 430, or when the remaining string is greater than or equal to the preset number, a notification message to the administrator terminal 230 By sending it, the administrator can select the model name 430 of the product.

For example, the negative keyword 431 may include 'Hz' indicating frequency, 'Kg' indicating weight, etc., 'Number + V' indicating voltage, and 'Number + W' indicating power. The form of can also be set as a negative keyword (431).

On the other hand, in another embodiment, after the data extraction unit 150 searches for a string including alphabets and numbers for all pages of the product manual, excludes the string containing the keyword 431 from among the searched strings. It is also possible to use a method of determining the model name 430 of the product from the remaining strings.

In addition, the data extraction unit 150 may extract structured data such as the manufacturer's phone number 440 and the manufacturer's homepage address 450 from the product manual. In addition, the data extraction unit 150 may store the structured data such as the extracted manufacturer's phone number 440 and the manufacturer's homepage address 450 in the storage unit 130 in association with the corresponding manual.

In general, phone numbers and homepage addresses have a certain format. For example, in the case of a phone number, it is written in the format of 'xx-xxx-xxxx', 'xxx-xxxx', 'xxxx-xxxx', and in the case of the homepage, 'www.xxxxx.com', 'www.xxxxx. co.kr '.

Therefore, the data extraction unit 150 may search for a character string that matches the phone number format and the homepage address format from the product manual, and extract it to the manufacturer's phone number 440 and the manufacturer's homepage address 450.

In addition, the data extracting unit 150 divides the range of the product manual to speed up the extraction of the manufacturer's phone number 440 and the manufacturer's homepage address 450, and the manufacturer's phone number 440 and the manufacturer's homepage address. The search for 450 may be performed sequentially.

For example, the data extraction unit 150 may first perform a search for the first and last pages of the product manual, which are likely to have the manufacturer's phone number 440 and the manufacturer's homepage address 450, If it is determined that the manufacturer's phone number 440 and / or the manufacturer's homepage address 450 does not exist after searching the corresponding page, then the remaining pages may be searched.

12 is a flowchart illustrating a method for providing a chatbot-based product manual according to an embodiment of the present invention.

Referring to FIG. 12, a method for providing a chatbot-based product manual according to an embodiment of the present invention may include an interactive interface providing step (S10), an answer content searching step (S20), and an answer content providing step (S30). have.

In the step of providing an interactive interface (S10), a chatbot-based interactive interface may be provided to the user.

For example, the interactive interface may provide a chat window for chatting between the chatbot and the user to the user terminal 220, through which the user can input a product related inquiry to be solved by the user.

In the answer content search step (S20), when product information and product-related inquiries are input from the user terminal 220 through an interactive interface, the answer information for the product-related inquiries may be searched by referring to the product information. .

In the response content providing step (S30), the response content retrieved in the response content searching step (S20) may be provided to the user through the chatbot in the interactive interface.

The interactive interface provision step (S10) and the answer content provision step (S30) may be performed through the chatbot service providing module 110 described above, and the answer content search step (S20) may be performed through the search module 120 described above. Can be performed.

13 is a flowchart illustrating an answer content search step according to an embodiment of the present invention.

13, the answer content search step (S20) may include a conversation history search step (S21), a product manual search step (S22), a question and answer information search step (S23), and an online search step (S24). .

In the conversation history search step (S21), the contents of an answer to a product-related inquiry can be searched for in the conversation history DB 131 that stores conversation contents made in the interactive interface.

In the product manual search step (S22), if an answer content for a product related inquiry is not searched in the conversation history search step (S21) or if there is an additional search request from the user terminal 220, a manual DB in which product manuals of multiple companies are stored At 132, a specific product manual corresponding to product information may be searched for, and an answer content for the product related inquiry may be searched for in the searched specific product manual.

In the question and answer information search step (S23), if a specific product manual corresponding to product information is not searched in the product manual search step (S22) or an additional search request is requested from the user terminal 220, the product information is referred to by referring to the product information. The contents of the answer to the matter can be searched in the question and answer DB 133.

In the online search step (S24), if the answer information for the product-related inquiry is not searched in the question-and-answer information search step (S23) or if there is an additional search request from the user terminal 220, the product-related inquiry is referred to the product information You can search for answers online.

On the other hand, the method for providing a chatbot-based product manual according to an embodiment of the present invention is a related keyword search step of searching for a product related keyword related to a corresponding product in the related keyword DB 134 using product information input through an interactive interface. It may further include.

Through this, in the interactive interface providing step (S10), the product related keyword searched in the related keyword search step may be provided to the user, and the product related keyword selected by the user among the product related keywords may be recognized as a product related inquiry. Can be. Accordingly, a service as shown in FIG. 7 can be provided to the user.

In addition, the method for providing a chatbot-based product manual according to an embodiment of the present invention may additionally include a manual collection step, a structured data extraction step, and a question and answer collection step.

In the manual collection step, a product manual provided from a company or an online product manual may be collected, which may be performed by the manual collection unit 140 described above.

In the structured data extraction step, structured data may be extracted from the product manual stored in the manual DB 132, which may be performed by the data extraction unit 150 described above.

In the question and answer collection step, the manufacturer's homepage address is accessed using the manufacturer's homepage address extracted in the standard data extraction step, and the question and answer information existing in the manufacturer's homepage can be collected and stored in the query response DB 133. This step may be performed by the Q & A collection unit 160 described above.

Those of ordinary skill in the art to which the present invention pertains will appreciate that the present invention may be implemented in other specific forms without changing its technical spirit or essential features. Therefore, it should be understood that the above-described embodiments are illustrative in all respects and not restrictive. The scope of the present invention is indicated by the scope of the claims, which will be described later, rather than the detailed description, and all the modified or modified forms derived from the meaning and scope of the claims and their equivalent concepts are included in the scope of the present invention. Should be interpreted.

100: chatbot-based product manual delivery system
110: chatbot service providing module 120: search module
130: database 131: conversation history DB
132: Manual DB 133: Q & A DB
134: associated keyword DB 135: product information DB
136: dialogue scenario DB 140: manual collection unit
150: data extraction unit 160: question and answer collection unit
200: terminal 210: enterprise terminal
220: user terminal
230: administrator terminal

Claims (22)

A chatbot service providing module that provides a user with a chatbot-based interactive interface, and provides answers to user's product related inquiries through the interactive interface;
A conversation history DB (database) for storing conversation content made in the interactive interface;
Manual DB that stores product manuals of multiple companies;
When product information and product-related inquiries are input from the user terminal through the interactive interface, the contents of the product-related inquiries are searched in the conversation history DB with reference to the product information, and the searched response contents are recalled. A search module provided as a chatbot service providing module;
A manual collection unit for collecting product manuals provided by a company or online product manuals and storing them in the manual DB; And
A data extraction unit for extracting structured data from the product manual stored in the manual DB; Including,
The product information includes a manufacturer of the product, a name of the product, and a model name of the product,
The chatbot service providing module, if any one of the manufacturer of the product, the name of the product and the model name of the product is input from the user terminal, the manufacturer of the product, the name of the product, and the Query for information that is not entered among product model names, and the query is made of options,
The structured data includes product-related keywords, the manufacturer of the product, the name of the product, the model name of the product, and the homepage address of the manufacturer,
The data extracting unit extracts a string including the manufacturer's estimated keyword from the product manual as the manufacturer of the product, and the manufacturer's estimated keyword includes 'stock company' or '(shares)',
The data extracting unit extracts a character string including English letters and numbers from the product manual as a model name of the product. According to claim 1,
The search module,
If the answer information for the product related inquiry is not searched in the conversation history DB or there is an additional search request from the user terminal, the specific product manual corresponding to the product information is searched in the manual DB, and the searched specific product manual A chatbot-based product manual providing system, characterized in that the product searches for answers to the product-related inquiries and provides them to the chatbot service providing module. According to claim 2,
The chatbot-based product manual providing system,
A Q & A DB that stores Q & A information collected from the manufacturer's homepage; Further comprising,
The search module,
When a specific product manual corresponding to the product information is not searched in the manual DB or an additional search request is requested from the user terminal, the query information DB searches for an answer to the product related inquiry by referring to the product information. And, the chatbot-based product manual providing system, characterized in that to provide the searched response content to the chatbot service providing module. According to claim 3,
The search module,
If the answer to the product-related inquiry is not searched in the query and answer DB, or if there is an additional search request from the user terminal, refer to the product information and search for the response to the product-related inquiry online. , Chatbot-based product manual providing system, characterized in that the searched answer is provided to the chatbot service providing module. delete delete According to claim 1,
The chatbot-based product manual providing system,
A related keyword DB in which product related keywords are stored for each product; And
The search module,
A chatbot-based product characterized in that a product related keyword related to a corresponding product is searched in the related keyword DB using the product information input through the interactive interface, and the searched product related keyword is provided to the chatbot service providing module. Manual delivery system. The method of claim 7,
The chatbot service providing module,
The product-related keyword provided from the search module is provided to the user through the chatbot, and the product-related keyword selected by the user among the product-related keywords is recognized as the product-related inquiry and transmitted to the search module. Product manual delivery system based on chatbot. delete delete According to claim 1,
A question-and-answer collection unit that accesses the manufacturer's homepage by using the manufacturer's homepage address extracted by the data extraction unit, and collects the question-and-answer information existing in the manufacturer's homepage and stores it in the query and response DB; Chatbot-based product manual providing system further comprising a. In a method for providing a chatbot-based product manual by a chatbot-based product manual providing system including a chatbot service providing module, a search module, a manual collection unit, and a data extraction unit,
The chatbot service providing module provides an interactive interface for providing a chatbot-based interactive interface to a user;
When the search module, product information and product-related inquiries are input from the user terminal through the interactive interface, an answer content search step of searching for answers to the product-related inquiries by referring to the product information; And
The chatbot service providing module, providing an answer content providing the searched answer content to the user through the interactive interface; Including,
The step of searching for the contents of the above answer,
Search the contents of answers to the product-related inquiries in the conversation history DB (database) that stores the conversations made in the interactive interface,
The product information includes a manufacturer of the product, a name of the product, and a model name of the product,
In the step of providing the interactive interface,
When any one of the manufacturer of the product, the name of the product, and the model name of the product is input from the user terminal, information about input information among the manufacturer of the product, the name of the product, and the model name of the product is input to the user. Perform a query, the query is made of options,
A manual collection step in which the manual collection unit collects a product manual provided by a company or an online product manual and stores it in a manual DB; And
The data extraction unit, the structured data extraction step of extracting the structured data from the product manual stored in the manual DB; Further comprising,
The structured data includes product-related keywords, the manufacturer of the product, the name of the product, the model name of the product, and the homepage address of the manufacturer,
In the step of extracting the structured data, the character string including the manufacturer's estimated keyword is extracted from the product manual as the manufacturer of the product, and the manufacturer's estimated keyword includes 'share company' or '(share)',
In the step of extracting the structured data, a method for providing a chatbot-based product manual is characterized in that a character string including alphabets and numbers is extracted from the product manual as a model name of the product. The method of claim 12,
The step of searching for the contents of the above answer,
If the answer to the product-related inquiry is not searched in the conversation history DB or if there is an additional search request from the user terminal, a specific product manual corresponding to the product information is searched for in the manual DB where product manuals of multiple companies are stored. And, in the searched specific product manual, the method for providing a product manual based on a chatbot, characterized in that it searches for an answer to the product related inquiries. The method of claim 13,
The step of searching for the contents of the above answer,
If a specific product manual corresponding to the product information is not searched in the manual DB or an additional search request is made from the user terminal, the answer information for the product-related inquiry is searched in the query and answer DB with reference to the product information. ,
The query response DB,
A method for providing a product manual based on a chatbot, characterized in that it stores information about questions and answers collected from the manufacturer's homepage of the product. The method of claim 14,
The step of searching for the contents of the above answer,
If the answer to the product-related inquiry is not searched in the query and answer DB, or if there is an additional search request from the user terminal, the answer information for the product-related inquiry is searched online by referring to the product information. Characterized in that the chatbot-based product manual providing method. delete delete The method of claim 15,
A retrieval keyword search step in which the search module searches for a product related keyword related to a corresponding product in a related keyword DB using product information input through the interactive interface; Further comprising,
The related keyword DB,
A method for providing a chatbot-based product manual, wherein product-related keywords are stored for each product. The method of claim 18,
In the step of providing the interactive interface,
A method for providing a chatbot-based product manual, characterized in that a product related keyword searched in the related keyword search step is provided to a user, and a product related keyword selected by the user among the product related keywords is recognized as the product related inquiry. delete delete The method of claim 12,
Collection of question-and-answers characterized by accessing the manufacturer's homepage using the manufacturer's homepage address extracted in the structured data extraction step, and collecting and answering the question-and-answer information existing in the manufacturer's homepage. step; Method for providing a chatbot-based product manual, further comprising a.

Images