KR20210100389A - Method for searching similar patents - Google Patents

Abstract

The present invention is to provide a method for searching for similar patents, which comprises the steps of: analyzing a target claim of a target patent input by a searcher; extracting candidate documents from a document database using a keyword extracted from the target claim; comparing the target claim and the candidate documents to determine the similarity of the target claim of each of the candidate documents; and providing a searcher terminal with a list of the candidate documents in order of the similarity. The step of displaying a list of the candidate documents on the searcher terminal in order of the similarity, includes a step of displaying a text of a segment related to the target claim in each of the candidate documents using at least one of a background color, a character color, and a font, which are distinguished from other segments. Accordingly, the accuracy of searching for similar documents is improved.

Description

Similar patent search method {METHOD FOR SEARCHING SIMILAR PATENTS}

The present invention relates to a similar patent search method, similar patent search software, and similar patent search system, and more particularly, to a similar patent search method, software and system capable of improving the accuracy and efficiency of similar prior patent search.

When a patent dispute occurs, finding prior art documents that can invalidate the disputed patent (hereinafter also referred to as "target patent") determines the success or failure of the dispute. Conventionally, when a searcher accesses a database that provides a prior art search service (for example, http://www.kipris.or.kr provides a patent document database) and inputs a search keyword for searching prior art documents, , a list of prior art documents (also referred to as “hit documents”) including the entered keyword is displayed on the display of the searcher computer. This list includes summaries of hit documents (eg, abstracts of patent publications) or representative drawings, and the searcher reviews information included in the list while scrolling through the display.

When a searcher finds documents that look similar to the target patent through summary or drawings, he or she must click on each document and review the contents in detail to check whether the information is actually meaningful. Therefore, when the number of hit documents is large (usually hundreds to thousands), it takes a lot of time to review the summary or drawings itself, and there is a risk that the searcher loses concentration and loses meaningful information. In addition, if a summary or representative drawing of a document does not provide sufficient information about the document, the accuracy of the search is severely reduced. Moreover, since it is necessary to return to the initial search screen and re-enter keywords for re-search, there are problems that the search takes a long time and the efficiency is low.

Therefore, there is a need for a search system that increases the accuracy of a similar literature search and enables a user to review it efficiently.

In order to solve the problems of the prior art as described above, the present invention comprises the steps of analyzing a target claim of a target patent input by a searcher; extracting candidate documents using the keyword extracted from the target claim; determining a similarity of each candidate document to the target claim by comparing the target claim with the candidate documents; and providing the list of candidate documents to the searcher terminal in the order of the degree of similarity, wherein the displaying of the list of candidate documents to the searcher terminal in the order of the degree of similarity includes, in each candidate document, the target claim It provides a similar patent search method, comprising the step of displaying the text of the segment related to the segment using at least one of a background color, a character color, and a font that is distinguished from other segments.

1 is a conceptual diagram of a similar patent search system according to an embodiment of the present invention.
2 is a block diagram conceptually illustrating the configuration of a similar patent search server according to an embodiment of the present invention.
3 is a flowchart illustrating a similar patent search method according to an embodiment of the present invention.

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

1 is a conceptual diagram of a patent search system according to an embodiment of the present invention. The patent search system 100 includes a searcher computer 10 that is a client and a patent search server 20 . The searcher computer 10 accesses the patent search server 20 through an Internet browser, and retrieves information input by the searcher through an input device (eg, keyboard, touchpad, voice input, etc.) to the patent search server 20 ) to transmit or receive information provided by the patent search server 20 and include hardware and software necessary to provide the information to a searcher through an output device (eg, a display device).

The patent search server 20 receives the number of the target patent entered by the searcher through the searcher computer 10 and the number of the claim to be searched (hereinafter, also referred to as "target claim"), and analyzes the target claim of the target patent. and similar prior literature. Previous documents searched in this way are called candidate documents. The patent search server 20 also analyzes the degree of similarity between the candidate documents and the target claim and provides candidate documents with high similarity to the searcher computer 10 .

In an embodiment, the patent search server 20 provides a list of prior documents (hereinafter, "similar documents") having a high degree of similarity to the target claim to the searcher computer 10 in an order of high similarity. In particular, the patent search server 20 identifies segments in each similar document that are relevant to the target claim and provides information about the segments to the user computer 20 . The searcher computer 10 receives the information and displays the relevant segment of the similar literature in a user identifiable form in the browser.

In this embodiment, it is assumed that the searcher computer 20 and the patent search server 10 are connected via the Internet, but the similar patent search system of the present invention is not limited to this embodiment. In another embodiment, software for performing a search function of the patent search server may be installed in the searcher computer in the form of an application.

2 is a block diagram conceptually illustrating the configuration of a patent search server 20 according to an embodiment of the present invention. The patent search server 20 according to this embodiment includes a web app and UI server 210 , a document DB 220 , a document data processing engine 230 , a target claims processing engine 240 , and a document analysis engine 250 . includes The web app and UI server 210 includes a search case management app/UI 212 and a result review app/UI 214 . The search case management app/UI 212 includes an application and a UI that allow a searcher to store and manage their searched cases and results. The result review app/UI 214 includes an application and UI for displaying a list of similar documents in the browser of the searcher's computer, and allowing the searcher to conveniently review the contents of each similar document. In an embodiment, the result review app/UI 214 provides a list of similar documents to the browser of the searcher's computer in the order of high similarity, and when the searcher selects a similar document to be reviewed, target claims in the similar document At least one of the backround color, character color, and font (font, size, thickness, etc.) of the text of segments related to is displayed differently from the rest of the text. For example, the text of the segment related to the target claim has a yellow background color, and the text of the remaining parts does not specify a background color, so that a searcher can easily identify the related segment.

In an embodiment, the web app and UI server 210 may further include a target claim editing app/UI (not shown). A target claim editing app/UI (not shown) allows a searcher to edit a target claim into one or multiple segments. For example, when a searcher divides a target claim into one or a plurality of segments (hereinafter, "claim segment") according to meaning, the patent search server 20 compares the claim segment with text segments of a prior document, thereby determining the target claim and The degree of similarity of each prior literature is judged. In this embodiment, the result review app/UI 214 may have a different method of displaying related segments of similar documents for each claim segment of the target claim. For example, if there are three claim segments, the segment of the similar document corresponding to claim segment 1 has the background color of the text in yellow, the segment of the similar document corresponding to claim segment 2 has the background color of the text in red, and The segment of the similar document corresponding to the segment 3 can easily identify the segment of the similar document corresponding thereto for each claim segment by displaying the background color of the text in blue.

In addition, the plurality of claim segments may be categorized according to meaning. In this case, for each category, a corresponding segment of a similar document may be displayed in the same or similar manner for a plurality of claim segments belonging to each category. For example, when there are three claim segments belonging to one category, the searcher can be selected from the category and the It may also be possible to identify claim segments at the same time.

In one embodiment, the results review app/UI 214 allows a searcher to select (eg, by clicking a select button) some or all of a plurality of claim segments of a target claim. Results review app/UI 214 displays corresponding segments of similar literature, only for the claim segment selected by the searcher.

In another embodiment, the results review app/UI 214 allows the searcher to designate one or more of the segments of similar literature as relevant to a particular claim segment, and store information about the segment specified by the searcher in a database. Save.

The document DB 220 includes the type of document (for example, whether it is a patent document or not, whether a patent document is a published application or a registered publication, etc.), and bibliographic information (in the case of a patent document, the title of the invention, application number, and filing date). , publication number, publication date, registration number, registration date, applicant, inventor, summary of invention, priority date), and contents of documents (in the case of patent documents, detailed description and text of claims) are stored. The document stored in the document DB may be, for example, a thesis, and the type is not particularly limited. The document data processing engine 230 processes the text of documents in a searchable form, and stores the text of each document in the document DB 220 as a semantic unit.

When the target claim processing engine 240 receives the claim number of the target patent to be searched by the searcher from the searcher computer, the target claim processing engine 240 detects the text of the claim from the document DB 220 and provides it to the searcher computer. The web app and UI server 210 may provide the searcher computer with an interface that allows the searcher to edit the target claim.

In one embodiment, the target claims processing engine 240 divides the target claim into one or more segments (claim segments) in semantic units and provides the claim segments to the searcher computer. The web app and UI server 210 may provide the searcher computer with an interface that allows the searcher to edit the segment of the target claim.

The literature analysis engine 250 analyzes the target claim or claim segment to extract keywords, and uses the keywords to extract candidate documents. Next, the degree of similarity between the extracted candidate documents and the target claim is determined. To determine the similarity, for example, one or more of techniques such as keyword appearance frequency and similarity determination by machine learning may be used. In one embodiment, the document analysis engine 250 identifies a segment related to a target claim (or claim segment) in a similar document with a high degree of similarity among candidate documents, and provides information about the segment to a searcher computer. As described above, the searcher computer receives the information and displays the relevant segment of the similar document on the display in a user identifiable form (eg, background color, text color, or font, etc.) in the browser.

3 is a flowchart illustrating a similar patent search method according to an embodiment of the present invention. In step 302, the target claim of the target patent input by the searcher is analyzed. In step 304, candidate documents are extracted using the keyword extracted from the target claim. In step 306, the target claim and the candidate documents are compared to determine the degree of similarity of each candidate document to the target claim. In step 308, the list of candidate documents is provided to the searcher terminal in the order of the degree of similarity. The step of displaying the list of candidate documents on the searcher terminal in the order of the degree of similarity includes selecting at least one of a background color, a character color, and a font to distinguish the text of the segment related to the target claim in each candidate document from other segments. is used to display (step 310).

What has been described above is only some embodiments of a similar patent search method and system, and the present invention is not limited to the above-described embodiments, and those of ordinary skill in the art to which the present invention pertains Various modifications and other embodiments of the present invention can be derived from the related drawings. Various modifications are possible without departing from the general concept and spirit of the invention as defined by the appended claims and their equivalents.

Claims (1)

analyzing a target claim of a target patent input by a searcher;
extracting candidate documents from a document database by using the keywords extracted from the target claims;
determining a similarity of each candidate document to the target claim by comparing the target claim with the candidate documents; and
providing the list of candidate documents in the order of the degree of similarity to the searcher terminal;
including,
The step of displaying the list of candidate documents on the searcher terminal in the order of the degree of similarity includes selecting at least one of a background color, a character color, and a font to distinguish the text of the segment related to the target claim in each candidate document from other segments. comprising the step of displaying using
Similar patent search methods.

Images