KR102594717B1 - Priority-centered selection document adoption system based on multiple search keywords and drive method of the Same - Google Patents

Abstract

The present invention discloses document management and big data processing technology. In other words, the priority-based selection document adoption system and driving method according to the multiple notation of search keywords according to an embodiment of the present invention determines the number of search keywords indicated on each page of the entire document file through a count screen AI (artificial intelligence) engine. In addition, when the mouse approaches one of the paragraphs described on the page that are sequentially arranged based on the number of search keywords revealed in all pages of the selected document file among all document files, any one of the above is zoomed in. Displays in enlarged and embossed three-dimensional form.

Description

Priority-centered selection document adoption system based on multiple search keywords and drive method of the Same}

The present invention relates to document management and big data processing technology. In particular, the number of search keywords indicated on each page of the entire document file is identified through a count screen AI (artificial intelligence) engine, and the total page of the selected document file among the entire document file is identified. When the mouse approaches one of the paragraphs described on the page, which are sequentially arranged based on the number of search keywords revealed in the search keyword, one of the paragraphs is zoomed in and displayed in an enlarged and embossed three-dimensional shape. This relates to a priority-based selection document adoption system and its operation method according to multiple notations.

In general, a search engine refers to an Internet site that helps users search for sites that contain the information they are looking for among the numerous sites on the Internet. There are hundreds of search engines on the Internet.

Although there are no officially established standards for classifying search engines, they are generally divided into topic-specific search engines and keyword-type search engines depending on the operation mode of the search engine. Keyword-type search engines are further divided into general keyword-type search engines and front-end search engines. -It is divided into front-end search engines and intelligent search engines.

However, not all search engines are clearly differentiated as mentioned above, and 'YAHOO (brand name)', a synonym for topical search engines, also supports keyword search, and most keyword-type search engines also provide topical search services. provided at the same time.

A thematic search engine refers to a search engine that provides a list of information on the Internet classified according to major topics such as society, culture, art, sports, and politics. It provides a list of various information corresponding to the topic. Therefore, it is also called a directory server, subject catalog, menu search, etc., and has the advantage of being able to easily access the information when used in situations where it is not possible to extract special keywords or key words to find information, but it is difficult to access the desired information. Since you have to go through several steps such as 'major category → middle category → small category → information you are looking for', there is a risk that if you set the wrong path in the middle, you may end up further away from what you are looking for.

A conventional, keyword-type search engine refers to one that has built its own database of the contents and URLs (homepage addresses) of homepages on the Internet. It has the advantage of being able to quickly find the information you want by entering just a few keywords (search terms). However, if you perform a keyword search in a situation where the correct keyword cannot be extracted, it has the disadvantage of resulting in wasting a lot of time by outputting incorrect results, and a front-end search engine is one that uses many search engines. This refers to what is collected on the screen, and does not build an information database using robots, spiders, crawlers, etc., but only collects and provides 'search term input windows' of various engines, and is an outpost for entering other search engines. It is called a front-end search engine, and not only can you use a variety of search engines on one screen without having to move around and search through multiple search engines, but it also allows you to search not only for HTML documents on the web, but also for public search engines. It has the advantage of having a very wide search area that can even find software, newsgroups, or academic documents, but since it does not have its own database, it cannot support all of the various search options that can be used in each search engine, and it cannot support all the various search options that can be used in each search engine. It has the disadvantage of revealing a certain amount of information in leading searches.

Recently, as Internet use has become increasingly active, many people are expressing their opinions on the Internet through media such as blogs and wikis. Additionally, when evaluating the value of specific information, there is a growing demand to refer to opinion information posted by others on the Internet.

For example, there are a variety of user opinions on the Internet, ranging from product reviews to movie reviews. These opinions of each user can also be used when general users want to see the opinions of other users before purchasing a product or watching a movie, and marketers, stock traders, etc. can use the various opinions of general users about each product or company. It can also be used when you want to know. In particular, general users have a strong tendency to read other users' reviews before purchasing a specific product and then purchase that product.

However, the opinions that exist on the Internet exist only on individual websites, so when users want to use such opinion information, there is the inconvenience of having to manually search through each and every website.

It is difficult for users to find all of these websites, and if they want to find other users' opinions through general search, web documents with opinions, web documents with positive opinions, and web documents with negative opinions are mixed, effectively different users. There is a problem where it is difficult to find their opinions.

To solve these problems, user opinion extraction technology is being actively researched in academia at home and abroad, and in the field of information retrieval, there has been significant development since the early 2000s, and various technologies are being researched.

However, existing information retrieval technologies simply provide searches based on information where keywords exist, and also provide more high-level searches based on positively/negatively evaluated content in documents or sentences in which each keyword appears. I can't stay. Recently, attempts have been made to apply user opinion extraction technology to information retrieval, but they are still limited to simply dividing positive and negative documents.

Korean Patent Registration Number: 10-0869624-0000 Korean Patent Registration Number: 10-2174462-0000 Korean Patent Registration Number: 10-0903726-0000 Korean Patent Registration Number: 10-0796905-0000

The priority-based selection document selection system and its operating method according to the multiple notation of search keywords of the present invention were devised to solve the problems of the prior art, and the first purpose of the present invention is to select The number of search keywords is determined through the count screen AI (artificial intelligence) engine, and the mouse moves the mouse to any one of the paragraphs described on the page that are sequentially arranged based on the number of search keywords revealed in all pages of the selected document file among all document files. When approaching, one of the above is zoomed in and displayed in an enlarged and embossed three-dimensional shape, thereby intensively analyzing key paragraphs that are highly related to the search keyword and key content mentioned in the detailed article. This is so that you can learn quickly and accurately.

In addition, the second purpose of the present invention is to clearly monitor file list information containing search keywords among a large number of document files by utilizing software-developed text reading and paragraph special demonstration content (web/app) and process them. This is to activate it as quantitative data or to use it as big data applied to opinion analysis or various statistical analyses.

In addition, the third purpose of the present invention is to crawl all pages opened by selecting a document file, detect pages containing search keywords that are words of interest and related words for all pages, and then provide information so that they can be viewed, so that even fields of interest are not covered. This is to reduce as much as possible the wasteful elements of having to spend time reading all of the pages described together, and to provide a selective reading information support service tailored to consumers to enable readers to think selectively and analyze efficiently.

The present invention for achieving the above problems includes the following configuration. In other words, the priority-based selection document adoption system according to the multiple notation of search keywords according to an embodiment of the present invention performs an entire folder search or a specified folder search for all document files stored in the database by entering a search keyword in the search box. As a result, at least one document file included in the entire folder or a designated folder is searched and the first letter of the at least one document file is identified to generate and display file list information automatically sorted in Korean consonant order or alphabet order. When the entire folder search or the specified folder search operation is performed, the number of search keywords displayed on each different page of the entire document file is automatically identified by a count screen AI (artificial intelligence) engine that counts and screens the search keyword among the entire document file. A file list information start section that sorts document files including the most frequently listed pages in descending or ascending order; And when one of the document files recognized from the file list information start unit is selected, the count screen AI engine is activated to screen all pages of the selected document file and read the text, and among all the pages whose text has been read, the The page with the most search keywords is given priority and is called sequentially, and as the count screen AI engine divides the detailed articles written on the called page into paragraphs and automatically displays them differently, among the classified paragraphs When the mouse approaches any one of them, it zooms in and displays it in an enlarged and embossed three-dimensional shape, and automatically assigns a background color to the search keywords indicated in any one of them to color correct them so that they stand out from the background. Includes text reading and paragraph special opening section.

In addition, the file list information starting unit according to an embodiment of the present invention enters a search keyword into a search box and performs an entire folder search or a designated folder search for all document files stored in the database unit; The file list information starting unit searches for at least one document file included in the entire folder or a designated folder and identifies the first letter of the at least one document file to generate file list information automatically sorted in Korean consonant order or alphabet order. step of showing; Automatically identifying, by the file list information starting unit, a count screen AI (artificial intelligence) engine that counts and screens the number of search keywords displayed on each different page of the entire document file when the entire folder search or designated folder search operation is performed; arranging, by the file list information starting unit, document files including pages on which the search keyword is most frequently described among all document files in descending or ascending order; selecting one of the document files recognized from the file list information start unit by a character reading and paragraph special opening unit; The character reading and paragraph special expression unit operates the count screen AI engine to screen all pages of the selected document file and read the characters, and gives priority to the page with the most search keywords among all the read text pages. sequentially calling steps; The count screen AI engine automatically displays detailed articles written on the called page separately by paragraph; When the character reading and paragraph special display unit approaches one of the classified paragraphs with a mouse, zooming in on one of the paragraphs and displaying it in an enlarged and embossed three-dimensional shape; And a step of automatically assigning a background color to the search keywords indicated in any one of the text reading and paragraph special expression units to perform color correction so that they stand out from the background.

The priority-based selection document adoption system and method for operating the same according to the multiple notation of search keywords of the present invention determines the number of search keywords indicated on each page of the entire document file through a count screen AI (artificial intelligence) engine, and determines the number of search keywords displayed on each page of the entire document file. When the mouse is brought close to any one of the paragraphs described on the page, which are sequentially arranged based on the number of search keywords revealed in all pages of the selected document file, any one of the paragraphs is zoomed in and displayed enlarged and embossed. By displaying it in a shape, it focuses on key paragraphs that are highly related to the search keyword and analyzes them, giving the first effect of quickly and accurately understanding the key content mentioned in the detailed article.

In addition, the present invention utilizes software-developed text reading and paragraph special demonstration content (web/app) to clearly monitor file list information containing search keywords among a large number of document files, process them, and activate them as quantitative data. It gives a second effect that is useful as big data that is applied to opinion analysis or various statistical analyses.

In addition, the present invention crawls all pages opened by selecting a document file, detects pages containing search keywords that are words of interest and related words for all pages, and provides information so that they can be viewed, so that fields of interest are also described. It reduces as much as possible the wasteful elements of having to spend time reading the entire page, and provides a third effect of providing a customized selective reading information support service to enable readers to think selectively and analyze efficiently.

Figure 1 is a diagram illustrating a priority-based selection document selection system based on multiple notation of search keywords according to an embodiment of the present invention.
Figure 2 is another diagram showing a priority-based selection document selection system according to multiple notation of search keywords according to an embodiment of the present invention.
Figure 3 is a diagram showing the main components of the priority-based selection document selection system based on multiple notation of search keywords according to an embodiment of the present invention implemented on content for character reading and paragraph special demonstration.
Figure 4 is a flowchart showing a method of operating a priority-based selection document selection system according to multiple notation of search keywords according to an embodiment of the present invention.

[Example]

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

Figure 1 is a diagram illustrating a priority-based selection document selection system based on multiple notation of search keywords according to an embodiment of the present invention.

Referring to FIG. 1, the priority-based selection document selection system 1000 according to the multiple notation of search keywords of the present invention uses a screen AI (artificial intelligence) engine 400 to count the number of search keywords indicated on each page of the entire document file. When the mouse approaches one of the paragraphs described on the page, which are sequentially arranged based on the number of search keywords revealed in all pages of the selected document file among all document files, the immediately recognized paragraph is displayed due to the proximity movement. It is a system that zooms in to enlarge and display in an embossed three-dimensional shape, including a file list information opening unit (100), a character reading and paragraph special opening unit (200), an automatic meta information display unit (300), and a count. Includes a screen AI (artificial intelligence) engine 400.

The priority-based selection document adoption system (1000) according to multiple notation of search keywords is a file list information start section (100), character reading and paragraph specialization content on text reading and paragraph special demonstration content developed by software to enable app/web access. Note that the special display unit 200, the automatic meta information display unit 300, and the count screen AI (artificial intelligence) engine 400 are mounted to support signal processing to implement the following operations.

More specifically, as can be seen in FIG. 3, the file list information initiating unit 100 enters a search keyword in the search box 110 to search the entire folder for all document files stored in the database unit (not shown). Performs a search of the specified folder.

The file list information initiator 100 searches for at least one document file included in the entire folder or a designated folder, and identifies the first letter of the at least one document file to provide file list information automatically sorted in Korean consonant order or alphabet order. Create it and show it.

In addition, when searching an entire folder or searching a designated folder, the file list information initiator 100 automatically uses the count screen AI (artificial intelligence) engine 400 to count and screen the number of search keywords displayed on each different page of the entire document file. Sort document files including pages with the most search keywords among all document files in descending or ascending order.

The file list information starting unit 100 may sort the meta information of at least one document file by sorting it by file connection program name, or rearrange it in descending or ascending order by file size, file creation date, or file download. It is rearranged in descending or ascending order by year, month, and date.

When the character reading and paragraph special opening unit 200 selects one of the document files recognized from the file list information opening unit 100, it operates the count screen AI engine 400 to screen all pages of the selected document file. It reads text and calls it sequentially, giving priority starting from the page with the most search keywords among all pages read .

In addition, the character reading and paragraph special display unit 200 automatically displays the detailed text written on the page on which the count screen AI engine 400 is called by dividing it into paragraphs, so that the mouse approaches any one of the classified paragraphs. When doing so, one of the paragraphs is zoomed in and displayed in an enlarged and embossed three-dimensional shape.

In addition, the character reading and paragraph special display unit 200 automatically assigns a background color to the search keywords indicated in one of the paragraphs and performs color correction using the count screen AI engine 400 to make it stand out from the background.

Upon completion of text reading, the text reading and paragraph special display unit 200 converts and creates document files that were limited for reading only into read/write document files for a fee or free of charge to customer communication terminals who wish to review detailed texts. After payment approval, information is provided or files are downloaded.

In addition, the character reading and paragraph special expression unit 200 generates a selected one of a Hancom Office Word file, an MS Office Word file, a Google Word file, and a text Word file that belong to read/write document files.

The character reading and paragraph special display unit 200 sets the screen to be displayed with different colors, different thicknesses, underlines, and different fonts for the search keywords that appear on each page where at least one search keyword is listed, so that the search keywords are displayed. We provide support so that the importance of is emphasized when reviewing the content.

In addition to the search keywords indicated on each page, the character reading and paragraph special expression unit 200 sets other search keywords that appear repeatedly at least three times as derived keywords and uses different colors, different thicknesses, underlines, and different fonts to differentiate them from the search keywords. Indicates distinction.

The character reading and paragraph special expression unit 200 operates the count screen AI engine 400 to count the number of search keywords expressed at least one on each page of the selected document file and sorts them in order of pages with the highest number of search keywords. Pages with the fewest search keywords are sorted in order, and when displaying the screen, the page number of the page currently implemented on the screen and the number of search keywords are displayed together.

In another embodiment according to the present invention, when the character reading and paragraph special display unit 200 clicks on any one of the paragraphs displayed separately with a mouse, the clicked one pops up and displays it to be visible in a pop-up window.

In addition, as the character reading and paragraph special expression unit 200 performs natural language processing (NLP) according to the operation of the count screen AI engine 400, the subject, object, and predicate existing in the sentence described in any clicked item are essential. Under the condition that it be used as an input, a short sentence (summary) is automatically generated that simply summarizes the sentences recognized from any one clicked.

As shown in FIG. 2, the automatic meta information display unit 300 displays the file name, brief description of the file, file connection program name, file author name, file size, file creation date, file download date, and file source that can be confirmed from the selected document file. The meta information, including name, number of pages, and file path, is displayed on the screen as is, or some of the meta information is selectively converted to binary, octal, or hexadecimal numbers and displayed on the screen to ensure information security and prevent information leakage.

Figure 4 is a flowchart showing a method of operating a priority-based selection document selection system according to multiple notation of search keywords according to an embodiment of the present invention.

Referring to Figure 4, the method of operating the priority-based selection document adoption system according to multiple notation of search keywords determines the number of search keywords indicated on each page of the entire document file through a count screen AI (artificial intelligence) engine, and When the mouse approaches one of the paragraphs described on the page, which are sequentially arranged based on the number of search keywords revealed in all pages of the selected document file among the files, any one of the above is zoomed in and displayed enlarged and embossed. This is a driving method that displays in a three-dimensional shape.

First, the file list information initiating unit enters a search keyword in the search box and performs an entire folder search or a designated folder search for all document files stored in the database unit (S100).

The file list information start section searches for at least one document file included in the entire folder or a designated folder and identifies the first letter of at least one document file to generate and display file list information automatically sorted in Korean consonant order or alphabet order ( S200).

When performing an entire folder search or a designated folder search operation, the file list information start unit automatically determines the number of search keywords displayed on each page of the entire document file using a count screen AI (artificial intelligence) engine that counts and screens (S300).

The file list information starting unit sorts document files including pages with the most search keywords among all document files in descending or ascending order (S400).

The character reading and paragraph special opening unit selects one of the document files recognized from the file list information opening unit (S500).

The character reading and paragraph special expression unit runs the count screen AI engine to screen all pages of the selected document file to read the characters, and calls them sequentially, starting from the page with the most search keywords among all read pages. S600).

The count screen AI engine divides detailed articles written on the called page into paragraphs and automatically displays them differently (S700).

When the character reading and paragraph special display unit approaches one of the marked paragraphs with the mouse, it zooms in on the paragraph immediately recognized through proximity and displays it in an enlarged and embossed three-dimensional shape (S800).

The character reading and paragraph special expression department automatically assigns a background color to the search keywords indicated in any one of them to make them stand out from the background. Color correction is performed using deep learning techniques or genetic algorithms based on (non-)reinforcement learning implemented in the count screen AI engine. Perform (S900).

Although the present invention has been described above with reference to preferred embodiments, those skilled in the art may make various modifications and changes to the present invention without departing from the spirit and scope of the present invention as set forth in the claims below. You will be able to understand that it exists.

Priority-based selection document adoption system based on multiple search keyword notation (1000)
File list information start section (100)
Character reading and literary specials (200)
Automatic meta information display unit (300)
Count Screen AI (400)

Claims (8)

By entering a search keyword in the search box and performing an entire folder search or a specified folder search for all document files stored in the database section, at least one document file included in the entire folder or a specified folder is searched and the at least one document file is searched. By identifying the first letter of the document file, it generates and displays file list information automatically sorted in Korean consonant order or alphabet order.
When searching the entire folder or searching the specified folder, the number of search keywords displayed on each different page of the entire document file is automatically determined by the count screen AI (artificial intelligence) engine, and the page on which the search keyword is listed the most among the entire document file is automatically identified. A file list information starting unit that sorts document files including in descending or ascending order; and
When selecting one of the document files recognized from the file list information start unit, the count screen AI engine is activated to screen all pages of the selected document file and read the text, and perform the search among all pages whose text has been read. It is called sequentially, giving priority starting from the page with the most keywords.
As the count screen AI engine automatically displays the detailed articles written on the called page separately by paragraph, when the mouse approaches one of the paragraphs, the mouse zooms in on one of the paragraphs. It displays the search keywords in an enlarged and embossed three-dimensional shape, and automatically assigns a background color to the search keywords indicated in any one of the above to perform color correction to make them stand out from the background. A priority-based selection document adoption system.
The method of claim 1, wherein the file list information starting unit,
Among the meta information of the at least one document file, it can be displayed sorted by file connection program name, rearranged in descending or ascending order by file size, and rearranged in descending or ascending order by file creation date or file download date. A priority-based selection document adoption system based on multiple notation of search keywords, characterized by showing.
The method of claim 1, wherein the character reading and paragraph special expression unit,
When the text reading is completed, the document file, which was limited to reading only, is converted and created into a read/write document file, and information is provided or the file is downloaded after approving paid or free payment to the customer's communication terminal who wishes to review the detailed text. I do it for you,
A priority-based selection document selection system based on multiple notation of search keywords, characterized in that the read/write document file is generated as a selected one of Hancom Office Word file, MS Office Word file, Google Word file, and text Word file.
According to claim 1,
The meta information including the file name, file brief description, file connection program name, file author name, file size, file creation date, file download date, file source name, number of pages, and file path that can be confirmed from the selected document file is displayed on the screen. Search keyword multiple search, characterized in that it further includes a meta information automatic display unit that displays or selectively converts some of the meta information into binary, octal, or hexadecimal numbers to display on the screen in order to ensure information security and prevent information leakage of the meta information. Priority-based selection document adoption system according to notation.
The method of claim 1, wherein the character reading and paragraph special expression unit,
The importance of the search keyword is emphasized when reviewing the content by setting the screen to be displayed in a different color, thickness, underline, and font for the search keyword that appears on each page where at least one search keyword is listed. A priority-based selection document adoption system based on multiple notation of search keywords, characterized by supporting this.
The method of claim 1, wherein the character reading and paragraph special expression unit,
A search characterized in that, in addition to the search keywords indicated on each page, other search keywords that appear repeatedly at least three times are set as derived keywords and displayed differently from the search keywords with different colors, different thicknesses, underlines, and different fonts. A priority-based selection document adoption system based on multiple keyword notation.
The method of claim 1, wherein the character reading and paragraph special expression unit,
When you click on any one of the separated paragraphs with the mouse, one of the paragraphs pops up and is displayed in a pop-up window,
As natural language processing is performed according to the operation of the count screen AI engine, a short sentence in which the sentence is briefly summarized is generated on the premise that the subject, object, and predicate existing in the sentence described in any of the above are required inputs. A priority-based selection document adoption system based on multiple notation of search keywords.
A step where the file list information initiating unit enters a search keyword in the search box and performs an entire folder search or a designated folder search for all document files stored in the database unit;
The file list information starting unit searches for at least one document file included in the entire folder or a designated folder and identifies the first letter of the at least one document file to generate file list information automatically sorted in Korean consonant order or alphabet order. step of showing;
automatically determining the number of search keywords displayed on each page of the entire document file using a count screen artificial intelligence (AI) engine when the file list information initiator operates to search the entire folder or search the specified folder;
arranging, by the file list information starting unit, document files including pages on which the search keyword is most frequently described among all document files in descending or ascending order;
selecting one of the document files recognized from the file list information start unit by a character reading and paragraph special opening unit;
The character reading and paragraph special expression unit operates the count screen AI engine to screen all pages of the selected document file and read the characters, and gives priority to the page with the most search keywords among all the read text pages. sequentially calling steps;
The count screen AI engine automatically displays detailed articles written on the called page separately by paragraph;
When the character reading and paragraph special display unit approaches any one of the classified paragraphs with a mouse, zooming in on one of the paragraphs and displaying it in an enlarged and embossed three-dimensional shape; and
A priority-based selection document selection system according to multiple notation of search keywords, including the step of the character reading and paragraph special expression unit automatically assigning a background color to the search keywords indicated in any one of the above and performing color correction to make them stand out from the background. How to drive.