KR102434880B1 - System for providing knowledge sharing service based on multimedia platform - Google Patents

Abstract

Provided is a knowledge sharing service providing system based on multimedia platforms, which comprises: a user terminal extracting search results by Wikipedia, Q&A, video sharing, and file sharing categories and separately displaying the search results by category when a search term is entered in an all search menu; and a sharing service providing server, which includes: a search unit providing search results corresponding to a search term by Wikipedia, Q&A, video sharing, and file sharing categories when the search term is entered in the all search menu on the user terminal; a Wikipedia unit providing a Wikipedia menu, allowing a document for at least one keyword to be created, edited, and modified by at least one participant terminal, and allowing search results corresponding to the search term to be extracted from a Wikipedia database when the search term is input from the user terminal; a Q&A unit providing a Q&A menu, when a question is uploaded from the user terminal and at least one answer is registered, allowing the user terminal to select the best answer, and when the user terminal searches for an answer to a question, extracting search results from a Q&A database by using the question as a search term and displaying the search results on the user terminal; a video sharing unit providing a video sharing menu and allowing videos to be uploaded or searched in a video sharing database; and a file sharing unit providing a file sharing menu and tools for creating, editing, and deleting files or folders, and setting registration, modification, and deletion of folder access rights. The knowledge sharing service providing system provides integrated search, Wikipedia, Q&A, video sharing, and file sharing services on an integrated platform, thereby briefly providing the search results without visiting each platform.

Description

Multimedia platform-based knowledge sharing service providing system {SYSTEM FOR PROVIDING KNOWLEDGE SHARING SERVICE BASED ON MULTIMEDIA PLATFORM}

The present invention relates to a system for providing a knowledge sharing service based on a multimedia platform, and provides a system capable of providing Wikipedia, Q&A, video sharing and file sharing in one site.

Implicit knowledge that can be created and built through the experiences and secrets of organizational members in corporate management activities, that is, knowledge sharing activities that increase problem-solving ability by storing, sharing, and acquiring experiential knowledge is a new core concept in the era of the 4th industrial revolution. positioned as a keyword. According to the internationalization, systemization, automation, and digitalization of business, the core talents of the 4th industrial age are changing into knowledge workers. is becoming the source of It shows that the source of economic wealth is shifting from tangible assets such as land, machinery, equipment, and hardware assets to intangible assets such as ideas and software that create core competencies of a company. In other words, it suggests that securing valuable knowledge and securing capabilities that can be used effectively are key factors for a company to secure competitiveness.

At this time, in order to share knowledge, a question-and-answer method or a method using a video has been researched and developed. In No. 0039585 (published on April 16, 2010), if the category and keyword of a question to be delivered from the knowledge sharing site are set, the question is input through the community site, and when the category and keyword match when providing an answer, knowledge sharing When the initial price of the answer request question is registered in order to share knowledge using the configuration linked to the site and the video, the answerer makes an answer video, and an advertisement video related to the answer request question is selected and uploaded along with the answer video, A configuration in which the interrogator receives and outputs this is disclosed, respectively.

However, among the above configurations, in the former case, there is no meaning beyond simply linking the question-and-answer site and the community site, so it is only a configuration to receive an alarm through the community site in an environment where the instant messenger is not activated. is changed to a video format, and does not constitute an integrated platform. Accordingly, research and development of a platform that can provide integrated search results by integrating various services such as Wikipedia for knowledge sharing, question answering, video sharing, and file sharing into one platform is required.

In one embodiment of the present invention, integrated search, Wikipedia, question and answer, video sharing, and file sharing service can be provided in one site, and the search results are provided on the integrated platform of the present invention, so that each platform is not visited. You can see the search results at a glance without the need to, and if the file or folder name and the search word match, the file or folder is searched. It is possible to provide a method for providing a system for providing a knowledge sharing service based on a multimedia platform that allows the user to move to content, and to search for a scene by detecting a frame corresponding to a scene in the video when a search word describing a scene in the video is input. However, the technical task to be achieved by the present embodiment is not limited to the above-described technical task, and other technical tasks may exist.

As a technical means for achieving the above-described technical problem, in one embodiment of the present invention, when a search word is input in the full search menu, the search results are extracted by Wikipedia, Q&A, video sharing, and file sharing categories and divided into categories. When a search term is entered in the full search menu in the displayed user terminal and the user terminal, a search unit that provides search results corresponding to the search term by categories of Wikipedia, Q&A, video sharing and file sharing, Wikipedia menu is provided, and at least Wikipedia, a Q&A department that allows a document for at least one keyword to be created, edited, and modified by one participant terminal, and a search result corresponding to the search word is extracted from the Wikipedia database when a search word is input from the user terminal When a menu is provided and a question is uploaded from the user terminal, when at least one answer is registered, the best answer is selected from the user terminal. It provides a question and answer section that extracts and displays a video sharing menu, a video sharing section that allows videos to be uploaded or searched in a video sharing database, and a file sharing menu that allows you to create, edit, and delete files or folders. It provides a tool and includes a sharing service providing server including a file sharing unit for registering, modifying and deleting access rights of folders.

According to any one of the above-described problem solving means of the present invention, integrated search, Wikipedia, Q&A, video sharing, and file sharing service can be provided in one site, and the search result is provided in the integrated platform of the present invention. You can see the search results at a glance without visiting each platform, and if the file or folder name matches the search term, the file or folder is searched. It is moved to multimedia content mapped with a unique code, and when a search word describing a scene in a video is input, a scene search is also possible by detecting a frame corresponding to a scene in the video.

1 is a diagram for explaining a system for providing a knowledge sharing service based on a multimedia platform according to an embodiment of the present invention.
FIG. 2 is a block diagram illustrating a shared service providing server included in the system of FIG. 1 .
3 and 4 are diagrams for explaining an embodiment in which a multimedia platform-based knowledge sharing service providing system according to an embodiment of the present invention is implemented.
5 is an operation flowchart illustrating a method of providing a system for providing a multimedia platform-based knowledge sharing service according to an embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily carry out 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.

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

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

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

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

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

Hereinafter, the present invention will be described in detail with reference to the accompanying drawings.

1 is a diagram for explaining a system for providing a knowledge sharing service based on a multimedia platform according to an embodiment of the present invention. Referring to FIG. 1 , the multimedia platform-based knowledge sharing service providing system 1 may include at least one user terminal 100 , a sharing service providing server 300 , and at least one participant terminal 400 . However, since the multimedia platform-based knowledge sharing service providing system 1 of FIG. 1 is only an embodiment of the present invention, the present invention is not limitedly interpreted through FIG. 1 .

At this time, each component of FIG. 1 is generally connected through a network (Network, 200). For example, as shown in FIG. 1 , at least one user terminal 100 may be connected to the shared service providing server 300 through the network 200 . In addition, the shared service providing server 300 may be connected to at least one user terminal 100 and at least one participant terminal 400 through the network 200 . In addition, at least one participant terminal 400 may be connected to the shared service providing server 300 through the network 200 .

Here, the network refers to a connection structure in which information exchange is possible between each node, such as a plurality of terminals and servers, and an example of such a network includes a local area network (LAN), a wide area network (WAN: Wide Area Network), the Internet (WWW: World Wide Web), wired and wireless data communication networks, telephone networks, wired and wireless television networks, and the like. Examples of wireless data communication networks include 3G, 4G, 5G, 3rd Generation Partnership Project (3GPP), 5th Generation Partnership Project (5GPP), Long Term Evolution (LTE), World Interoperability for Microwave Access (WIMAX), Wi-Fi (Wi-Fi) , Internet, LAN (Local Area Network), Wireless LAN (Wireless Local Area Network), WAN (Wide Area Network), PAN (Personal Area Network), RF (Radio Frequency), Bluetooth (Bluetooth) network, NFC ( Near-Field Communication) networks, satellite broadcast networks, analog broadcast networks, Digital Multimedia Broadcasting (DMB) networks, and the like are included, but are not limited thereto.

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

The at least one user terminal 100 may be a terminal that inputs a search word using a multimedia platform-based knowledge sharing service providing system related web page, an app page, a program, or an application and outputs the result. In this case, the user terminal 100 may be a terminal for uploading content corresponding to the material to be searched, and in this case, may be a terminal receiving points or rewards suitable for a preset reward system. In addition, the user terminal 100 may be a terminal that is granted access to the content to be shared when it is included in a rank or organization corresponding to the disclosure range of the content to be shared. In this case, the user terminal 100 may be a terminal of a general user, but may also be a terminal of an employee belonging to a company or organization and using an in-house knowledge sharing platform. However, the same composition may be applied to the two subjects except for the disclosure scope, so the subject should not be limited or restricted. Of course, it is self-evident that general users can use the functions of the public scope, and this does not exclude this function from general users.

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

The sharing service providing server 300 may be a server that provides a multimedia platform-based knowledge sharing service providing system web page, an app page, a program, or an application. In addition, the sharing service providing server 300 may be a server providing a search menu, a Wikipedia menu, a question and answer menu, and a video sharing menu. In this case, the shared service providing server 300 may build a database for each menu separately, but does not exclude the integration of the database. In addition, the shared service providing server 300 may be a server that provides a related tag for a search result and displays the number of times each related tag is inquired or searched. In addition, the shared service providing server 300 collects search logs from at least one tester terminal (not shown), receives a designation from the start point to the end point of the search, and maps a search process for a chain search in which a bundle of search terms appears ( It may be a server that builds a Map). When this map is built, the server 300 for providing shared services to users who enter the same search term thereafter provides information about what results other people have received through what search terms in the form of a mind map, so that the user can quickly find the search term desired by the user. It could be a server that helps you reach it. In addition, the sharing service providing server 300 may be a server that provides video sharing, Wikipedia, and Q&A menus, and brings multi-platform search results and divides them by multi-platform when providing an integrated search. In this case, the multi-platform may be, for example, Google, Naver, YouTube, etc., but is not limited thereto.

Here, the shared service providing server 300 may be implemented as a computer capable of accessing a remote server or terminal through a network. Here, the computer may include, for example, navigation, a laptop equipped with a web browser, a desktop, and a laptop.

The at least one participant terminal 400 may be a terminal of a participant who creates a Wikipedia using a web page, an app page, a program, or an application related to a multimedia platform-based knowledge sharing service providing system. Of course, the user of the user terminal 100 may also be a participant in creating Wikipedia.

Here, the at least one participant terminal 400 may be implemented as a computer that can access a remote server or terminal through a network. Here, the computer may include, for example, navigation, a laptop equipped with a web browser, a desktop, and a laptop. In this case, the at least one participant terminal 400 may be implemented as a terminal capable of accessing a remote server or terminal through a network. The at least one participant terminal 400 is, for example, a wireless communication device that guarantees portability and mobility, including navigation, Personal Communication System (PCS), Global System for Mobile communications (GSM), Personal Digital Cellular (PDC), PHS(Personal Handyphone System), PDA(Personal Digital Assistant), IMT(International Mobile Telecommunication)-2000, CDMA(Code Division Multiple Access)-2000, W-CDMA(W-Code Division Multiple Access), Wibro(Wireless Broadband Internet) ) terminal, a smartphone, a smart pad, a tablet PC, etc. may include all kinds of handheld-based wireless communication devices.

2 is a block diagram illustrating a shared service providing server included in the system of FIG. 1, and FIGS. 3 and 4 are an embodiment in which a multimedia platform-based knowledge sharing service providing system according to an embodiment of the present invention is implemented. It is a drawing for explaining.

Referring to FIG. 2 , the sharing service providing server 300 includes a search unit 310 , a question answering unit 320 , a video sharing unit 330 , a Wikipedia unit 340 , a setting unit 350 , and related tags. Unit 360, target limiting unit 370, display unit 380, bookmark unit 390, search log analysis unit 391, automatic correction unit 393, file sharing unit 394, code setting unit 395 ), an artificial intelligence search unit 396 , a request unit 397 , and a copyright infringement prevention unit 398 .

The shared service providing server 300 according to an embodiment of the present invention or another server (not shown) operating in conjunction with at least one user terminal 100 and at least one participant terminal 400 shares knowledge based on a multimedia platform. When transmitting a service providing system application, program, app page, web page, etc., at least one user terminal 100 and at least one participant terminal 400, a multimedia platform-based knowledge sharing service providing system application, program, app You can install or open pages, web pages, etc. In addition, the service program may be driven in at least one user terminal 100 and at least one participant terminal 400 by using a script executed in a web browser. Here, the web browser is a program that enables the use of a web (WWW: World Wide Web) service, and refers to a program that receives and displays hypertext written in HTML (Hyper Text Mark-up Language), for example, Netscape. , Explorer, Chrome, and the like. In addition, the application means an application on the terminal, for example, includes an app (App) executed in a mobile terminal (smartphone).

Referring to FIG. 2 , the search unit 310 may provide a search menu and classify search results by at least one platform to display them in the user terminal 100 . In this case, when a search word is input in the search menu, the user terminal 100 may extract a search result for the search word from at least one platform and display it separately for each at least one platform. For example, assuming that at least one platform is Naver's Knowledge IN, Wikipedia, and YouTube, search results for a search term may be displayed separately for each platform, that is, by Knowledge IN, Wikipedia, and YouTube.

The question answering unit 320 provides a question answering menu and when at least one answer is registered when a question is uploaded from the user terminal 100, the user terminal 100 selects the best answer, and the user terminal 100 In the case of searching for an answer to a question, the search result may be extracted from the question and answer database as a search word and displayed on the user terminal 100 . The user terminal 100 may upload a question from the question answer menu or search for and output an answer to the question.

The video sharing unit 330 may provide a video sharing menu and allow the video to be uploaded or searched for in a video sharing database. The user terminal 100 may upload or search for a video in the video sharing menu and output it. At this time, the video sharing unit 330 additionally extracts the meta tag using the meta tag extracted with the dictionary and the rule and the neural network model to find the video (content) that meets the user's intention, so that it more meets the user's needs A hybrid media content search query analyzer that can provide content Media content query analyzer, first, extracts keywords from input sentences and categorizes meta-types for keywords, secondly, searches for meta tags most similar to the extracted keywords, and thirdly, extracts keywords from the dictionary and rules. Through three steps of performing intelligent merging of meta tags and additionally extracted meta tags, it is possible to provide content by interpreting meta tags that are not defined in dictionaries and rules.

For example, content search means providing the corresponding content by extracting meta information from the input sentence to the content. In this case, the meta information may be composed of meta types such as genre, performer, and emotion. The meta type is, for example, a set of meta tags such as <Genre: Drama>, <Performer: Share>, and <Emotion: Emotional>. to be. Since users freely input what they want, they can input meta tags with a modified vocabulary. For example, the <emotion: emotional> meta tag can be morphologically changed like the words “impressed”, “feeling”, and “impressed”, and synonymous words such as “impressive” and “emotional” can be entered as The transformed vocabulary is difficult to recognize as a meta tag, and there may be no corresponding content, so a process of converting it into a meta tag is necessary.

At this time, keyword extraction from input sentences and meta-type classification for keywords is a sequence labeling problem that sets a label for each input vocabulary, such as entity name recognition. In the recent object name recognition task, the models using RoBERTa among Pretrained Language Models show good performance, and the model that recognizes the object name through the LSTM layer and CRF layer with the output of RoBERTa shows excellent performance. Since content is defined as information about meta tags, it is not possible to obtain desired results when searching using keywords. To solve this problem, a meta tag search process for the content most similar to the extracted keyword is required. At this time, research and methods such as constructing pairs of natural language sentences and synonym sentences in order to connect the object names that connect words and related documents using a neural network model and research that allow searching for synonyms and synonyms by constructing a dictionary Developed, these methods require large amounts of data. In addition, as new vocabulary is added every day, it is very expensive to build a thesaurus and thesaurus, and when using a neural network model, it is difficult to immediately apply a new meta tag because it takes time to learn the model.

Accordingly, in an embodiment of the present invention, a hybrid media content search query analyzer can be used, which extracts meta information using a dictionary and a rule, and extracts meta tags that cannot be extracted with a dictionary and a rule using deep learning. It allows us to provide content that is more tailored to the user's intentions. Meta information extraction using dictionaries and rules can extract meta tags defined in manually built dictionaries and rules. In this case, since the dictionary and the rule use morpheme information, meta tag information defined in the dictionary and the pattern can be extracted after morpheme analysis is performed on the input sentence. Dictionaries and rules have the advantage that they can be applied immediately without a learning process, so new meta tags can be applied for immediate interpretation, and the interpretation quality of the input, which was well interpreted only with the dictionary and rules, can be maintained.

Keyword extraction from input sentences and meta-type classification for keywords enable the extraction of keyword types including NONE tags for each vocabulary. To clarify the scope of extracted keywords, IOB tags are used to classify tags. Since the method used in entity name recognition can be used, in an embodiment of the present invention, a model in which an LSTM layer and a CRF layer are added to the RoBERTa result can be used. In addition, meaningful information extracted through morpheme analysis information, keyword dictionary, and patterns is used as an additional input to the LSTM layer. In order to search for a meta tag of content most similar to the extracted keyword, synonyms may be extracted without constructing a large amount of data. For example, it is possible to construct a meta tag searcher using a method using the same vocabulary, keywords such as substrings, and morphological information of meta tags, and weight information for vocabulary using BM25 scores. Finally, intelligent merging of meta tags extracted with dictionary and rules and additionally extracted meta tags can be solved by creating rules. Merge rules are designed to improve keywords such as resolving ambiguity between dictionary and pattern-based keywords and deep learning keywords, removing noise on keywords, and converting <Genre: Romance> <Genre: Comedy> into higher-order keywords such as <Genre: Romance Comedy>. rules that can be You can use the merged keywords to create queries that retrieve media content.

For example, if a query is generated after extracting meta tags for an input of "Emotional Romance Drama", the dictionary and pattern-based meta tag extraction for the input "Emotional Romance Drama" is <Genre: Romance> <Content: Drama> Extract meta tags. The LSTM Layer receives RoBERTa as a result of word vector, morphological analysis information, and meta tag information using a dictionary and pattern, and delivers it to the CRF Layer after calculation. The CRF Layer can additionally recognize <Emotional: Impressive> that was not recognized in the dictionary or pattern by inputting the value calculated by the LSTM Layer. <Emotional: Impressive> can be replaced with the corresponding meta tag as the meta tag search result is most similar to <Emotional: Impressive>, and after merging with dictionary and pattern-based keywords, a query to search media content can be created.

The Wikipedia unit 340 provides a Wikipedia menu, allows a document for at least one keyword to be created, edited, and modified by the at least one participant terminal 400 , and inputs a search term in the user terminal 100 . In this case, the search result corresponding to the search term can be extracted from the Wikipedia database.

The setting unit 350 may set an access authority group, designate a department for each access authority group, and designate a reward for multimedia data uploaded to the knowledge sharing service providing server 300 as a point.

The related tag unit 360 may extract a related tag for a search word input from the user terminal 100 and output it together on a screen including the search result. When the related tag is extracted and output, the related tag unit 360 outputs the number of times the related tag is searched with the related tag, and the related tag is sorted by at least one sort filter including alphabetical order and newest order. have. In this case, the related tag is a keyword of the related search word as the concept of the following query expansion.

In the field of information retrieval, search term recommendations or suggestions are mostly closely related to query expansion. Query expansion is a technique for refining a user's initial query. It is performed by adding a search term or phrase to the user's query, and is executed automatically by the search system or manually by the user in the form of relevance feedback. While the traditional query expansion technique is usually based on the analysis of words appearing in documents, the search word recommendation method in the web search field uses various factors such as the results frequently searched by users among the search results for the query as well as the words appearing in the document. There is a difference in that related search terms are extracted based on

The method of extracting and suggesting suitable search terms from the query log of a web search engine is to use the method of recommending search terms that appear together in the session in which a specific query is searched as related search words, or, in the case of a specialized search engine, the query log of a specific field After extracting search words continuously input by a user within a specific session as a related search word candidate group, an algorithm that recommends search words having a high content similarity among search words included in this candidate group as a related search word may be used. Alternatively, an algorithm (Snippet Click Models) that recommends words appearing in a snippet of a document searched by users after entering a specific query as related search words may be used. Alternatively, a modified Apriori algorithm (association analysis) may be used to find the correlation between the collected search terms. In addition to the large set search words searched for more than a certain number of times by the data mining algorithm, search terms searched for less than a certain number of times. It is also possible to use a method for finding the associative law of a small set. Alternatively, to support query expansion, terms extracted from web documents visited by search engine users may be recommended as related search terms, and the usefulness of the collected search terms may be ranked. In addition, a modified TF-IDF (Term Frequency-Inverse Document Frequency) method may be used as a method for extracting correlation between keywords, and various methods may be used.

When a document for at least one keyword is uploaded by the at least one participant terminal 400 , the target limiting unit 370 , the company disclosure (L1), the division disclosure (L2), the headquarters disclosure (L3), and the team disclosure (L4), part disclosure (L5) and responsible disclosure (L6) can be selected for disclosure within the organization. This may be a function used when a certain company or organization needs to limit what should be disclosed. Of course, even if it is used by an individual, it is as described above that the object that can be limited is not excluded.

The display unit 380 may extract and display two lines of a text related to a link, a title, a date of creation, and a keyword when providing a search result in the search menu. At this time, when the display unit 380 clicks or touches the search result in the user terminal 100 and selects it, it goes to the corresponding page and shows this screen. Since the screen always comes out from the top, from the first page, the user works hard. find the search term. If the page of the link is short, you can find it right away, but one page is long and you have to scroll for a long time, because it is a document file, so you have to search from the first page, or press Ctrl+F to find the data you are looking for. do. Accordingly, the display unit 380 increases convenience by using functions such as scrolling down or scrolling up or page down/up directly to the area where the search word exists, and highlighting the search word part and immediately showing it as soon as it goes to the link. can

For example, a user entered a search term "The Lord of the Rings: The Two Towers music analysis", and the search format was .pdf. At this time, a number of .pdf links are output as search results, and the user has to open each pdf file and repeat the process of finding the keyword "The Lord of the Rings: The Two Towers". It would not be very inconvenient if only one document was found and ended, but if it is a research process as one of the processes for writing a thesis to be submitted to a journal, it is necessary to repeat at least tens or hundreds of clicks and keyword searches. Even if you put it in and search for it, it may not be the material you want, so time and effort is wasted considerably. Accordingly, if you mouse over the text 2 line area containing keywords on the search result screen, the page containing the keyword found with Ctrl+F is overlaid on the search result page so that the user can edit the pdf one by one. You can avoid having to repeat the process of opening and finding. If one page does not have to go to the link of the search result, the search activity can be completed quickly and easily even in a situation where there is a delay due to insufficient networking resources or a delay due to insufficient computing resources.

The bookmark unit 390 may classify and display at least one search result bookmarked by the user terminal 100 into search, Q&A, video, and Wikipedia, respectively.

The search log analysis unit 391 displays a start point and an end point of a search from at least one tester terminal (not shown), and receives a search word cluster that is a time-series search word mass input up to the start point and end point of the search, and the search word is A single search in which only one is inputted and a chain search generating at least one search word cluster may be distinguished, and at least one search word cluster included in the chain search may be classified and stored into a category tag, a navigation tag, and an instance tag. Referring to FIG. 4 and Table 1 below, there are a single search, which is terminated at once by inserting an accurate search term, and a chain search, in which a search term is refined or interests are refined.

division Contents example single search Only one search term occurs in a series of search processes #oriental casablanca Chain Search:
Query cluster occurrence Refine search terms Correct spelling or form #Casablanca->#Casablanca refinement of interests Deepening or shifting the topic #big white lily->#big white lily variety->#oriental variety->#oriental Casablanca

For example, user A, who sees a large white lily on the way home from work, wants to buy lily bulbs in spring. However, user A does not know the name or variety of the large white lily. At this time, user A searches with the search term "big white lily". At this time, User A is a lily he passed by, so he moves to the image search tab based on what he saw and finds a blog with similar images. And user A goes to the blog and sees it, but there are a lot of words about how pretty it is, and the breed is not listed. Again, user A realizes that he has not entered the keyword (search term) of cultivar, and searches for "big white lily variety".

At this time, after seeing that oriental lilies are large and fragrant in the search results, search for "oriental lily types" again. Only then will the search results appear in the Lily/Lower Classifications tab with various varieties and names. At this time, change the search result to the image tab again and find the lily of the image similar to the one you saw. When I clicked on a link with an image, the image I saw did not appear immediately. I scrolled down to see the image I saw, but it did not appear until I rolled the mouse ball more than 10 times. Then the user gets the result [Oriental: Casablanca]. In this process, the search log becomes [Large and white lily (start point) -> Large white lily varieties -> Oriental lily type -> Oriental: Casablanca (end point)] The tag is [Large and White Lily Varieties->Oriental Lily Type], and the instance tag is [Oriental: Casablanca].

At this time, if these logs are collected and analyzed and then clustered through data mining, when user B enters the search term [big white lily], this process is shown as a mind map, and the result is immediately [ Oriental: Casablanca]. Initially, a large number of participants or beta testers must be recruited to create a dataset to learn through the process of specifying and labeling these search logs one by one by the participants and testers. By automatically analyzing the search logs of users who use the site of

In this case, text mining is a technology that aims to extract and process useful information from non/semi-structured text data based on natural language processing technology. Through text mining technology, it is possible to extract meaningful information from a large collection of texts, identify links with other information, find categories of texts, or obtain results beyond simple information search. Using this, large-capacity language resources and statistical and regular algorithms can be used to analyze identifiers or natural language input as queries, and to discover hidden information therein. In addition, cluster analysis may be used to finally discover a group of similar characteristics while merging objects having similar characteristics. These groups can be classified through cluster analysis, and when a new search word is inputted after identifying the characteristics of the cluster, objects according to the new search word can be classified.

In this case, the search word cluster (clusters) may be a series of search word clusters displayed by testers, participants, or users themselves. From this, it can be divided into a single search in which only one search word appears from the time the search action starts and ends until the end of the search action, and a chain search in which multiple search word bundles are generated. In addition, in the chain search, it can be classified into search word refinement, which corrects spelling or the shape of a search word, and interest refinement, which shows deepening or shifting of a topic. As a result of classifying the search word used in the chain search through scoring along the axis of 'categorical attribute-detailed attribute', it can be divided into three types as shown in FIG. 4(b). For example, things with large categorical attributes such as #interior and #art museum can be classified as category tags, and things with large detailed attributes such as #corduroy pants and #fleece can be classified as instance tags. A navigation tag is a tag composed of a coined word of two or more words, and is a tag that plays an intermediate role between a category tag and an instance tag. For example, in the case of a search that starts with a category tag as shown in (c) of FIG. 4, for example, sequential refinement appears by moving from high-category tags such as #fashion and #fashion to navigation tags and instance tags. . On the other hand, in the case of a type starting with a navigation tag, starting with a navigation tag such as #ootd and #dailylook from the beginning, the final instance tag can be reached by going back and forth between the navigation tag and the instance tag. Also, in both types, search term refinement of navigation tags is prominent.

Using this, when a search word for starting a chain search is input from the user terminal 100 , the automatic correction unit 393 may transmit a search result of the search word included in the end point of the chain search.

The file sharing unit 394 provides a file sharing menu, provides tools for creating, editing, and deleting files or folders, and can be set to register, modify, and delete access rights of a folder. The file sharing unit 394 generates a file sharing code when the number of file sharing days is input while sharing files in the user terminal 100, maps the file sharing days and the file, and stores the file sharing code. If is entered, it can be processed as file sharing completed. The file sharing unit 394 may set a file or folder including the same file name or folder name as the search word input by the user terminal 100 to be output in the file sharing category when the search result of the full search menu is provided. .

The code setting unit 395 assigns a unique code to the multimedia content uploaded to Wikipedia, Q&A, video sharing, and file sharing, maps and stores, and when the unique code is selected, the multimedia content mapped with the unique code can be set to move to

The artificial intelligence search unit 396 may extract a scene of a video including a search word by using at least one artificial intelligence algorithm and provide it as a search result of a video sharing menu. In the field of artificial intelligence, the field of vision, which uses video and images, and the field of natural language processing are combined, for example, video/image Q&A, video/image subtitle creation, video search, video behavior recognition, video scene search. and the like can be given. Among them, the video scene search is a field that finds the temporal location of a scene corresponding to the description in a video by using a description of a certain scene as an input (Query) value, and can be applied to various fields. Accordingly, in an embodiment of the present invention, a model capable of searching for a scene within a video group by combining a video search and a video scene search is used. When a description of a specific scene in one video is given as a search term, a video containing the description is found, and then the temporal position corresponding to the description is specified in the video. The effectiveness of learning can be maximized by using contrast learning, which learns through comparison between correct and incorrect answers for a given scene.

<Search for video>

In the video search, learning is carried out in groups of videos and corresponding descriptions. B videos and corresponding B scene descriptions are given. At this time, the similarity between each video and each scene description is measured as a score. Learning proceeds so that the similarity score between the matching video and the description is high, and the similarity score between the video and the description that is not matched is low. By comparing matched and unmatched pairs rather than simply finding the correct answer, the model can learn more accurate representations. For this purpose, Contrastive Loss can be used. In this case, contrastive learning refers to learning so that the feature vector of an image matching the current image is closer, and the feature vector is farther away from the current image and other data.

<Search for video scene>

A video scene search creates several candidate scenes within a video. Each candidate scene is made in a variety of ways, from short to long scenes. Thereafter, the similarity between each scene and the given description is measured as a score, and learning proceeds so that the similarity score of the scene corresponding to the corresponding description increases.

<Search for scenes in a video group>

Since the video and the corresponding description are given as the correct answer, the scene search within the video group is not applied during learning, but only applied during inference. While measuring the similarity score between the description and each video, a scene candidate group is created within each video, and the similarity score between each candidate group and the video is measured. Thereafter, the final similarity score of each scene is calculated as the product of the video similarity score and the scene similarity score. For each video, features can be extracted frame by frame using ResNet and SlowFast learned in advance. A moving picture having N frames is composed of N D-dimensional vectors. A scene description is composed of a D-dimensional query vector through LSTM (Long Short-Term Memory) using each word as an input value.

The similarity score between each frame and the description is calculated as the product between the frame vector and the query vector as shown in Equation 2, and the similarity score Sv of the video V and the similarity score Sp of the candidate group P correspond to Equations 3 and 4 as shown in Equation 3 and Equation 4 It is calculated as the average of the similarity scores of the frames.

ps of the candidate group P is the position of the start frame of the candidate group, and pe indicates the position of the last frame of the candidate group. In the case of a video search, the correct answer is selected in the order of the highest Sv among several videos, and in the case of a video scene search, the correct answer is selected in the order of the highest Sp among multiple candidates within the same video. In the case of a scene search within a video group, a correct answer is selected from multiple videos and candidate groups of each video, and is calculated as a product of a video similarity score and a candidate group similarity score as in Equation 5.

The IoU between the selected candidate group and the correct answer is calculated through the final similarity score Svp, and if it exceeds the standard, the correct answer is calculated. In order to verify the performance of the model, the video scene search dataset can use the DiDeMo benchmark dataset, and in measuring each performance, if the IoU between the predicted scene and the correct scene exceeds X, can Also, if it is ranked within the N rankings in the order of the highest similarity, it can be measured as a correct answer.

The requesting unit 397 receives the education level of the user terminal 100 , measures a readability index of an answer to a question in the question answer menu, and determines the readability index of the user terminal 100 . If the index corresponding to the level of education is exceeded, rewriting can be requested from the participant terminal who provided the answer. In this case, when measuring the ototoxicity index, Dale-Chall's ototoxicity formula can be used.

The readability formula consists of a semantic variable and a syntactic variable in common. The former is a lexical variable, which is measured by the length or difficulty of the vocabulary, and the latter is a sentence-level variable, mostly sentence length. measured with The readability formula of Korean text considers only the difficulty of vocabulary as a variable at the lexical dimension. for judging. The Dale-Chall formula is a representative formula for measuring the difficulty of a text based on the length of a sentence and difficult vocabulary. Dale-Chall's readability formula R(d) for document d is as Equation 6 below.

In Equation 6, words is the number of words used in a given document, difficult words is the number of difficult words, and sentences is the number of sentences used in the document. In the right side of Equation 6, the front part means the ratio of difficult words, and the rear part defines the sentence length and is the number of words used per sentence. To compare the readability of answers to questions, we will use the Dale-Chall formula, which is known to have high validity while evenly reflecting lexical and sentence factors.

In order to use the Dale-Chall formula in Korean text, the number of words and criteria for identifying difficult words are needed. In addition, it should be possible to grasp how many sentences a given document consists of, and for this, it must be possible to distinguish sentences. Documents created on the web have a lot of inscriptions or grammatical errors, so there is a difficulty in that it is impossible to simply classify sentences based on punctuation marks such as periods or question marks. In order to distinguish sentences, preprocessing is essential. Here, removing stopwords or analyzing morphemes for tokenization or part-of-speech tagging is a well-known technology of the Natural Language Process (NLP), so a detailed description will be omitted. To count the number of words in a sentence, you must be able to distinguish them individually. In English, words can be divided by spacing, but unlike English, in Korean, words with practical meaning or elements with grammatical functions are combined between stems, so it is not appropriate to simply classify words by spacing. Therefore, it is necessary to classify parts having practical meaning into words and to determine the part-of-speech of the word corresponding to this part. In one embodiment of the present invention, in order to classify words and grasp the number of words, nouns, pronouns, verbs, adjectives, adjectives, adverbs, interjections, Use parts of speech words such as dependent nouns, auxiliary verbs, and auxiliary adjectives.

In order to distinguish difficult words from easy words, in English, words not in the list are classified as difficult words based on the Dale-Chall list of easy words. Influenced by this influence in the study on the official development of readability of Korean texts, a vocabulary list is used. The vocabulary for Korean education is implemented in the Basic Korean Dictionary of the National Institute of the Korean Language, and a total of 10,635 words are counted: 1,835 for beginner, 3,855 for intermediate, and 4,945 for advanced. classified by level. Beginner is a basic vocabulary that can be used in everyday and familiar situations, and Intermediate is an expanded basic vocabulary in everyday and social situations. Advanced consists of social, professional, and unfamiliar vocabulary. The index calculated by the Dale-Chall readability formula can be converted to grade level and used. For example, if the Dale-Chall index is 10 or higher, it is judged as a college graduate or higher, the index corresponding to one grade is 0.5, and the higher the readability index, the more difficult the text can be interpreted.

The copyright infringement prevention unit 398, when at least one participant terminal 400 attempts to upload multimedia contents to any one of Wikipedia, Q&A, video sharing, and file sharing menus, infringes copyright of multimedia contents. It is determined whether the ID of the registered copyright holder and the participant of the multimedia content are not the same, and the multimedia content to be uploaded is the same as the previously stored multimedia content. The corresponding unique code can be forcibly inserted and uploaded. In the Supreme Court decision 2016da271608 case, ① the illegality of copyrighted infringing posts, which is required for an online service provider (OSP) operating an internet portal site, to be recognized ② Receive a request to delete or block specific and individual posts from the victim (or even if you have not received a direct request from the victim, you were aware of the circumstances in which the post was posted or the existence of the post It should be clearly apparent from the outside), ③ It suggests three requirements that management and control of the postings will be possible technically and economically.

The person who takes the primary responsibility for postings that infringe copyright on the Internet is the person who directly posted the postings. Even if the above postings are posted on the Internet posting space provided by OSP and Internet users can easily find the postings through the search function, OSP cannot be held liable for copyright infringement on the basis of such circumstances. For copyright infringement using the Internet, it is natural that the direct infringer who transmits another person's work without permission is liable. The issue of imposing an obligation to prevent copyright infringement is emerging as an important issue.

Accordingly, as described above, each Wikipedia, Q&A, video sharing, and file sharing documents, pictures, texts, photos, files, folders, videos, etc. are mapped with unique codes and stored. At this time, all multimedia contents that are initially uploaded are stored with a unique code mapped, and even at this time, it is possible to search for the existence of a copyright and attach a unique code. If the unique code is registered, when uploading the same article, photo, image, file, or video as the same article, photo, image, file, or video, insert the unique code forcibly to indicate the copyright holder or source.

At this time, in the case of a moving picture, it becomes more difficult to find the same moving picture. Finding an ROI (Region of Interest) or preparing each frame is a method that increases the amount of computation and consumes a lot of hardware resources. Accordingly, in one embodiment of the present invention, in order to determine whether a video of the video sharing menu infringes copyright, when the number of people included in a frame constituting the video is equal to or greater than a preset number, a frame in which a preset number of people appears is set as a reference frame, a block is created by tying a preset number of adjacent frames based on the reference frame, and skeleton information of an object in the block is extracted using the OpenPose algorithm. In the case of the same skeleton information of the given moving picture, the unique code of the given moving picture may be forcibly inserted.

A reference point is needed to limit the skeleton information to a specific frame without capturing the entire frame of the image. Therefore, in one embodiment of the present invention, the number of people present in a frame is taken as a reference point and adjacent frames within an error range of 20% are bundled together to generate a block. For example, assuming that the number of people in the first frame is set as 12, if there are between 10 and 14 people in subsequent frames, they are grouped into a single block. When creating a key point block, the top 5 are created based on the block with the largest average number of people, but if there are many people in one frame, all of the key points can be filled with 1 when changing the key points to binary coordinates. When the feature point block of the image to be compared with the input image is filled with 1, the accuracy may decrease and the amount of computation may increase. Therefore, when generating a feature point block, the maximum number of people is limited to prevent all information from being filled with 1 when converting to binary coordinates. In addition, it is possible to reduce the calculation process, so that the efficiency can be improved. However, if only the number of people is limited, the accuracy is lowered, and if the number of frames is large, the calculation process is increased. Therefore, it is necessary to limit the number of frames in a single block. In order to increase efficiency, the maximum number of 15 people and 40 frames when generating a feature point block may be limited, but this may be updated to increase or decrease according to the test result.

With respect to a photo or an image, the ambiguity of copyright grant can be judged based on the degree of similarity. By establishing similarity analysis modeling using deep learning, it prevents the granting of copyrights to imitated works and allows copyrights and unique codes to be granted only to works that have passed the similarity check. In order to generate a unique code for a work, the similarity is first analyzed. Similarity analysis uses deep learning CNN model. A unique code is issued only for works that have passed the similarity check, otherwise, the unique code of a photo or image for which a unique code has already been issued is forcibly inserted. To build a deep learning model for detecting similar designs, the VGG-16 model and AlexNet model, which are modified from CNN models, can be used to extract features of images and analyze similarity. By referring to the image analysis model, it is possible to build a model that detects the features of an image that is transformed from a CNN and a fully convolutional neural network (FNN) model. Analyze the similarity of each image feature information in the built CNN model. In order to quantify the similarity of image feature information, a Spearman correlation coefficient may be used. After training the constructed CNN model as a training set, various techniques can be applied in the direction of increasing the accuracy.

Hereinafter, an operation process according to the configuration of the shared service providing server of FIG. 2 will be described in detail with reference to FIGS. 3 and 4 as an example. However, it will be apparent that the embodiment is only one of various embodiments of the present invention and is not limited thereto.

3 shows a menu and a flowchart of a homepage according to an embodiment of the present invention. Each menu has five main menus, namely, the main screen (integrated search), Wikipedia, Q&A, video sharing, and system settings when moving to the login screen and the knowledge sharing center as shown in FIG. 3A. The administrator can set the reward in FIG. 3F, and the rest are all user screens. Figure 3b is a menu that can search Wikipedia, Q&A, and video sharing DB at a time for the full search (integrated search). In My Document Box, bookmarks and articles written by me may be classified and displayed, and bookmarks may also be output by dividing them by platform (Wikipedia/Q&A/Video sharing). 3C is a Wikipedia menu, in which writing or searching for a written article is possible, a tag list and a list of posts related thereto can be displayed, and a search can also be performed in the most recent order. At this time, you can also use the save bookmark function, write a comment or edit a post. Referring to FIG. 3D , it is a question and answer menu, and it is possible to write a question or write an answer, or search for it. 3e is a video sharing menu, and similarly, it may be configured to write a video post, search, display a tag, or view a list of recent entries. 3F and 3G are file sharing menus, which are largely divided into folder management and file management, and in the case of folder management, create a folder, change a folder name, copy, cut, delete a folder, or copy internal files and folders , cut, delete, etc., grant or revoke the right to access the folder, and register, modify, or delete the folder access right. Also, in file management, you can copy, cut, rename, and delete files, upload and download files, limit the number of days to share when sharing files, set sharing codes, and share codes. If is entered, file sharing is treated as complete. Figure 3h is a system setting, it is possible to set an access authority group, it is possible to designate a department for each group, it is also possible to set a point.

The matters not described with respect to the method for providing a system for providing a system for providing a knowledge sharing service based on a multimedia platform of FIGS. 2 to 4 are the same as or described for the method for providing a system for providing a system for providing a knowledge sharing service based on a multimedia platform through FIG. 1 above. Since it can be easily inferred from the contents, the following description will be omitted.

5 is a diagram illustrating a process in which data is transmitted/received between components included in the multimedia platform-based knowledge sharing service providing system of FIG. 1 according to an embodiment of the present invention. Hereinafter, an example of a process in which data is transmitted/received between each component will be described with reference to FIG. 5, but the present application is not limited to such an embodiment, and the example shown in FIG. 5 according to the various embodiments described above will be described. It is apparent to those skilled in the art that the data transmission/reception process may be changed.

Referring to FIG. 5 , the sharing service providing server provides search results corresponding to the search word by Wikipedia, Q&A, video sharing, and file sharing categories when a search word is input from the user terminal in the full search menu (S5100).

And, the shared service providing server provides a Wikipedia menu, allows a document for at least one keyword to be created, edited and modified by at least one participant terminal, The search result is extracted from the Wikipedia database (S5200).

In addition, the shared service providing server provides a question answer menu and when a question is uploaded from the user terminal, when at least one answer is registered, the best answer is selected from the user terminal, and when an answer to the question is retrieved from the user terminal The search result is extracted from the question answering database as a search word and displayed in the user terminal (S5300).

In addition, the sharing service providing server provides a video sharing menu, uploads or searches a video to a video sharing database (S5400), provides a file sharing menu, and provides tools for creating, editing and deleting files or folders, and folders set to register, modify, and delete access rights of (S5500).

The order between the above-described steps ( S5100 to S5500 ) is merely an example and is not limited thereto. That is, the order between the above-described steps ( S5100 to S5500 ) may be mutually changed, and some of the steps may be simultaneously executed or deleted.

Matters not described with respect to the method for providing a system for providing a system for providing a knowledge sharing service based on a multimedia platform of FIG. 5 are the same as or described for the method for providing a system for providing a system for providing a knowledge sharing service based on a multimedia platform through FIGS. 1 to 4 above. Since it can be easily inferred from the contents, the following description will be omitted.

The method for providing a system for providing a multimedia platform-based knowledge sharing service according to an embodiment described with reference to FIG. 5 is also implemented in the form of a recording medium including instructions executable by a computer, such as an application or program module executed by a computer. can be Computer-readable media can be any available media that can be accessed by a computer and includes both volatile and nonvolatile media, removable and non-removable media. Also, computer-readable media may include all computer storage media. Computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data.

In the method for providing a system for providing a multimedia platform-based knowledge sharing service according to an embodiment of the present invention described above, an application basically installed in a terminal (which may include a program included in a platform or an operating system basically installed in the terminal) may be executed by the application store server, an application or an application (ie, a program) directly installed in the master terminal through an application providing server such as a web server related to the application or the corresponding service by the user. In this sense, the method for providing a system for providing a knowledge sharing service based on a multimedia platform according to an embodiment of the present invention described above is implemented as an application (that is, a program) installed basically in a terminal or directly installed by a user, and is installed on a computer such as a terminal. It can be recorded on a readable recording medium.

The foregoing description of the present invention is for illustration, and those of ordinary skill in the art to which the present invention pertains can understand that it can be easily modified into other specific forms without changing the technical spirit or essential features of the present invention. will be. Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive. For example, each component described as a single type may be implemented in a distributed manner, and likewise components described as distributed may also be implemented in a combined form.

The scope of the present invention is indicated by the following claims rather than the above detailed description, and all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts should be interpreted as being included in the scope of the present invention. do.

Claims (8)

a user terminal that extracts the search results for each category of Wikipedia, Q&A, video sharing, and file sharing when a search word is input in the full search menu, and divides and displays the search results by category; and
When a search term is entered in the full search menu in the user terminal, a search unit that provides search results corresponding to the search term for each category of Wikipedia, Q&A, video sharing, and file sharing, a Wikipedia menu is provided, and at least one A Wikipedia section that allows a document for at least one keyword to be created, edited, and modified by a participant terminal of If an answer menu is provided and a question is uploaded from the user terminal, the user terminal selects the best answer when at least one answer is registered, and when an answer to a question is retrieved from the user terminal, the question is asked as a keyword It provides a question answering unit that extracts search results from the answer database and displays it on the user terminal, a video sharing unit that provides a video sharing menu, and a video sharing unit that allows videos to be uploaded or searched in the video sharing database, a file sharing menu, and A file sharing unit that provides tools for creation, editing, and deletion, registering, modifying and deleting access rights of the folder, extracting a tag related to a search word input from the user terminal, and displaying the search result on a screen A related tag unit for outputting together, a display unit for extracting and displaying two lines of a link, title, date of creation and keyword related text when providing the search result, multimedia uploaded to Wikipedia, Q&A, video sharing and file sharing A shared service providing server including a code setting unit for assigning a unique code to the content, mapping and storing, and setting the unique code to be moved to the mapped multimedia content when the unique code is selected; includes,
The associated tag unit,
When the related tag is extracted and output, the number of times the related tag is searched is output together with the related tag, and the related tag is arranged to be sorted by at least one sort filter including alphabetical order and newest order Multimedia platform-based knowledge sharing service provision system.
The method of claim 1,
The file sharing unit,
When the number of file sharing days is input while performing file sharing in the user terminal, a file sharing code is generated and the number of file sharing days and the file are mapped and stored, and when the file sharing code is input, file sharing is completed Multimedia platform-based knowledge sharing service providing system, characterized in that the processing.
The method of claim 1,
The file sharing unit,
Multimedia platform-based knowledge sharing, characterized in that a file or folder including the same file name or folder name as the search word input by the user terminal is set to be output in the file sharing category when the search result of the full search menu is provided service delivery system.
delete The method of claim 1,
an artificial intelligence search unit for extracting a scene of a video including the search word by using at least one artificial intelligence algorithm and providing it as a search result of the video sharing menu;
Multimedia platform-based knowledge sharing service providing system, characterized in that it further comprises.
The method of claim 1,
The shared service providing server,
Receive the education level of the user terminal, measure the readability index of the answer to the question in the question answer menu, and the readability index of the answer exceeds the index corresponding to the education level of the user terminal a request unit for requesting rewriting to the participant terminal that provided the answer in one case;
further comprising,
When measuring the readability index, a system for providing a knowledge sharing service based on a multimedia platform, characterized in that Dale-Chall's readability formula is used.
The method of claim 1,
The shared service providing server,
When the at least one participant terminal attempts to upload multimedia content to any one of Wikipedia, Q&A, video sharing, and file sharing menus, whether the multimedia content is copyright infringement or not, It is determined that the ID of the copyright holder and the participant is not the same, the multimedia content to be uploaded is the same as the previously stored multimedia content, and when the ID is not the same and the multimedia content is the same, a unique code corresponding to the copyright source Copyright infringement prevention unit forcibly inserted and uploaded;
Multimedia platform-based knowledge sharing service providing system, characterized in that it further comprises.
8. The method of claim 7,
The copyright infringement prevention department,
In order to determine whether the video of the video sharing menu is infringing on copyright, if the number of people included in the frame constituting the video is equal to or greater than a preset number, a frame in which the preset number of people appears is set as a reference frame, and the A block is created by tying a preset number of adjacent frames based on the reference frame, and after extracting the skeleton information of the object in the block using the OpenPose algorithm, the skeleton information is a skeleton of a moving picture to which a unique code is given. If the information is the same, the multimedia platform-based knowledge sharing service providing system, characterized in that forcibly inserting the unique code of the video to which the unique code has been given.

Images