200915107 九、發明說明： 【發明所屬之技術領域】 本發明是關於一種檢索系統及方法，尤指一種應用於 專利全文上的檢索索引建立系統及方法。 【先前技術】 隨著科技的蓬勃發展與網路資訊的日益普及，各種文 檔亦與日倶增，使得傳統式的檢索習慣受到相當大的影響 與衝擊。因為這些巨量的資料若使用傳統的方式檢索，其 數量是相當驚人的，而且資料的利用率也不高。所以隨著 數位時代的來臨，發展對文檔的全文檢索成為了必然的趨 勢。 與此相應的各種檢索系統也因應而生，這些檢索系統 類似一個資料庫，能存儲大量的資料，滿足使用者的檢索 需求。在這些資料庫當中，往往存在一些相關攔位，這些 攔位可能存儲一些文字資訊。因此，當使用者在檢索資料 時，該使用者實際是對資料庫中的資料欄位進行檢索。然 而，使用者仍然需要耗費大量的精力與時間，將文檔的文 字資料錄入到資料庫中，從而給資料收集及後續的檢索造 成不便，從而影響到資料的利用價值。 【發明内容】 基於上述内容，有必要提供一種專利全文檢索的索引 建立系統，該系統包括藉由網路相連接的伺服器及客戶 端。該伺服器包括啟動模組、轉檔模組、索引建立模組及 資料庫。所述啟動模組，用於偵測資料庫是否增加了新的 7 200915107 專利文檔，如果資料庫中增加了新的專利文檔，則啟動所 述轉檔模組；所述轉檔模組，用於讀取該專利文檔的各組 成部分的内容，將該專利文檔的各組成部分内容以預先定 義的文檔格式進行保存，以完成轉檔；所述索引建立模組， 用於使用索引建立技術將上述轉檔後的專利文檔的各組成 部分内容添加至所述資料庫中，並對該資料庫中轉檔後的 所有專利文檔的各個組成部分的内容分別重新建立全文索 引。 所述客戶端包括檢索模組，用於根據用戶的檢索標的 在建立好索引的資料庫中進行專利全文檢索，並計算檢索 結果的關聯值，所述關聯值是檢索結果與檢索標的之間的 相似度；顯示模組，用於對上述檢索結果按照關聯值排序， 依序顯示與當前檢索標的相關的專利文檔的簡要資訊及顯 示用戶所選擇的專利文檔的全文資訊。 本發明還提供一種專利全文檢索的索引建立方法，該 方法包括步驟：當資料庫中增加了新的專利文檔時，讀取 該專利文檔的各部分内容；將該專利文檔的各部分内容以 預先定義的文檔格式進行保存，以完成轉檔；使用索引建 立技術添加該轉檔後的專利文檔各部分内容到資料庫中， 並對該資料庫中轉檔後的所有專利文檔的各個組成部分的 内容分別重新建立全文索引。 該方法還包括步驟：根據用戶的檢索標的在該建立索 引的資料庫中進行專利全文檢索並計算關聯值，所述關聯 值是檢索結果與檢索標的之間的相似度；對上述檢索結果 8 200915107 按照關聯值排序，依序顯示與當前檢索標的相關的專利文 檔資訊；及顯示用戶所選擇的專利文檔的全文資訊。 本發明可當伺服器中新增專利文檔時，自動檢測文檔 内容，將專利文檔各組成部分的内容，自動讀取出來，建 立索引並存儲到資料庫中。方便用戶進行專利文檔全文内 容的檢索，獲取與當前檢索關鍵字相關的專利文檔，並將 相關檢索結果按關聯程度排列顯示，方便用戶閱讀、參考。 【實施方式】 如圖1所示，是本發明專利全文檢索的索引建立系統 較佳實施方式的模組圖。該系統包括伺服器1和客戶端2， 二者藉由網路3連接。所述伺服器1包括資料庫17、啟動模 組11、轉檔模組12及索引建立模組13。所述客戶端2包括檢 索模組20及顯示模組21。 在本較佳實施方式中，每一專利文檔包括說明書摘 要、說明書、申請專利範圍三個組成部分，說明書部分又 分為標題、發明所屬之技術領域、先前技術、發明内容、實 施方式、圖式簡單說明等多個子部分。說明書摘要部分提 供相應的說明書摘要主要内容和關鍵字，同樣，說明書各 子部分提供相應子部分的主要内容和關鍵字，申請專利範 圍部分的每一項權利要求提供相應權利要求的主要内容和 關鍵字。 啟動模組11用於偵測資料庫17中是否增加了新的專利 文檔，當偵測到資料庫中增加了新的專利文檔時，啟動轉 檔模組12進行轉檔作業。轉檔模組12用於藉由文檔傳輸協 9 200915107200915107 IX. INSTRUCTIONS: [Technical Field] The present invention relates to a retrieval system and method, and more particularly to a retrieval index establishing system and method applied to the entire patent. [Prior Art] With the rapid development of technology and the increasing popularity of online information, various documents have also increased in popularity, which has caused considerable impact on traditional search habits. Because these huge amounts of data are retrieved in the traditional way, the number is quite amazing, and the utilization of the data is not high. Therefore, with the advent of the digital era, the development of full-text search of documents has become an inevitable trend. Corresponding various retrieval systems have also been created. These retrieval systems are similar to a database, which can store a large amount of data to meet the user's search needs. In these databases, there are often some related blocks, which may store some text information. Therefore, when the user retrieves the data, the user actually retrieves the data field in the database. However, the user still needs a lot of effort and time to input the document data into the database, which inconveniences the data collection and subsequent retrieval, thereby affecting the value of the data. SUMMARY OF THE INVENTION Based on the above, it is necessary to provide an index establishment system for patent full-text retrieval, which includes a server and a client connected by a network. The server includes a startup module, a transition module, an index creation module, and a database. The startup module is configured to detect whether a new 7 200915107 patent document is added to the database, and if a new patent document is added to the database, the conversion module is activated; Reading the contents of each component of the patent document, storing the components of the patent document in a predefined document format to complete the conversion; the index building module for using indexing technology The components of the patent document after the above-mentioned translation are added to the database, and the full-text index is newly re-established separately for the contents of each component of the patent document after the conversion in the database. The client includes a retrieval module for performing a patent full-text search in the database in which the index is established according to the retrieval target of the user, and calculating an associated value of the retrieval result, where the correlation value is between the retrieval result and the retrieval target The similarity degree display module is configured to sort the search results according to the associated values, sequentially display the brief information of the patent documents related to the current search target, and display the full-text information of the patent documents selected by the user. The invention also provides an index establishing method for patent full-text search, the method comprising the steps of: when a new patent document is added to the database, reading the contents of each part of the patent document; The defined document format is saved to complete the conversion; the index creation technique is used to add the contents of the translated patent document to the database, and the components of all the patent documents after the conversion in the database are The content is re-established separately. The method further includes the steps of: performing a full-featured patent search in the indexed database according to the user's search target and calculating a correlation value, wherein the associated value is a similarity between the search result and the search target; and the search result 8 200915107 Sorting according to the associated values, sequentially displaying the patent document information related to the current search target; and displaying the full-text information of the patent document selected by the user. The invention can automatically detect the content of the document when the patent document is added in the server, automatically read out the contents of each component of the patent document, and build an index and store it in the database. It is convenient for users to search the full text of patent documents, obtain patent documents related to the current search keywords, and display relevant search results according to the degree of association, which is convenient for users to read and refer to. [Embodiment] FIG. 1 is a block diagram showing a preferred embodiment of an index establishing system for a full-text search of the present invention. The system includes a server 1 and a client 2, both of which are connected by a network 3. The server 1 includes a database 17, a startup module 11, a transition module 12, and an index creation module 13. The client 2 includes a search module 20 and a display module 21. In the preferred embodiment, each patent document includes three parts of the specification abstract, the specification, and the patent application scope, and the specification part is further divided into a title, a technical field to which the invention belongs, a prior art, an inventive content, an implementation manner, and a graphic form. Briefly explain multiple subsections. The main part of the manual provides the main content and keywords of the corresponding manual. Similarly, the sub-sections of the manual provide the main content and keywords of the corresponding sub-parts. Each claim in the patent application scope provides the main content and key of the corresponding claims. word. The startup module 11 is configured to detect whether a new patent document is added in the database 17. When a new patent document is added to the database, the conversion module 12 is activated to perform a conversion operation. The transfer module 12 is used for transmission by the document 9 200915107
的專利文槽，依據上述描述的專利文權的組成部分對其^ :分析’讀取各組成部分的内容，並將該各組成部分的内 f以預先定義的文槽格歧行保存。賴先定義的文槽格 2文槽由標題、朗書摘要、說明書、申請專利範圍等 ，分組成’該預先定義的文槽格式的文槽所屬的類型可以 疋網頁、XML或其他可以轉換成文字槽案的類型。索引建 立換組13使用索引建立技術將上述轉槽之後的專利文析各 組成部分的内容添加到資料庫17中，並對該資料庫17中田的 所有轉槽之後的專利讀的各餘成部分的内容分別重新 建立全文索引’包括對標題建立全文索引， 分建立全文㈣，對制㈣讀立全文μ及對 =範圍部分建立全文㈣。所職立全文索引是指藉由^ 部分内容的每—個詞，對每—個詞建立—個索引， 指明該詞在每—部分出現的次數及位置。該資料庫 含與專利文播轉槽後的文财各組成部分相對應的搁位， 用來存儲專利文檔轉檔後各部分的内容和關鍵字。 檢索模組20料根制戶的檢索標的對上述建立 索引的資料庫17進行檢索。藉由檢帝每一 其中檢索標的為用戶輪入的檢索關鍵字，相 檢索標的在資料庫17中檢索出的所有專利 ，並計算關聯值。 檢索結果為按照The patent essay, according to the components of the patent literary rights described above, analyzes the contents of each component and stores the inner f of the components in a predefined stencil. The text slot defined by Lai Xian is divided into the title, the summary of the book, the specification, the scope of the patent application, etc., and the type of the pre-defined stencil format can be converted into a web page, XML or other. The type of text slot. The index creation change group 13 adds the contents of the components of the patent analysis after the above-mentioned transfer to the database 17 using the index creation technology, and the remaining portions of the patent read after all the slots of the database 17 The contents of the re-establishment of the full-text index respectively include the establishment of full-text indexing of the title, the establishment of the full text (4), the establishment of the full text of the (4) reading full text and the right part of the scope (4). Full-text indexing refers to the establishment of an index for each word by means of each word of the part of the content, indicating the number and location of the word in each part. The database contains the corresponding positions corresponding to the various components of the text after the transfer of the patent text, and is used to store the contents and keywords of the parts after the patent document is transferred. The retrieval module 20 searches for the index database 17 of the above-mentioned index. By examining each of the search keywords in which the target is rounded up by the user, all patents retrieved in the database 17 are retrieved and the associated values are calculated. The search results are as follows
部分的條_難，…：！：料文擒各組成 列文槽的簡要資 號等，關聯值為 200915107 目包括專利文檔的標題、說明書摘要、說明書内容及申請 專利範圍。檢索模組20進行的檢索方式可為單一條目檢索 和多條目聯合檢索。顯示模組21用於對上述檢索結果依關 聯值排序，依序顯示上述檢索出的與當前檢索標的相關的 專利文槽的簡要資訊’從而方便閱讀、參考。該顯示模組 21還用於當用戶點擊檢索出的專利標題時將專利文檔下載 或開啟供用戶閱讀。該顯示模組21還用於在下載專利全文 前進行專利全文的預覽。 如圖2所示，是本發明專利全文檢索的索引建立方法 的車父佳實施方式的流程圖。首先，啟動模組n在偵測到資 料庫中增加了新的專利文檔時，啟動轉檔模組12藉由FTp 從資料庫17中獲取該專利文檔（步驟S2〇)。轉標模組12讀 取該專利文樓各組成部分的内容，並將讀取的專利文樓的 各組成部分的内容以預先定義的文檔格式進行保存，該預 先定義的文檔格式的文檔所屬的類型可以是網頁、XML或 其他可以轉換成文字檔案的類型（步驟S21)。索引建立模 纽13使用索引建立技術將上述轉檔之後的專利文檔的各部 分内容添加到資料庫17中，並對該資料庫17中的轉檔之後 的所有專利文檔的各個組成部分的内容分別重新建立全文 索引，包括對標題建立全文索引，說明書摘要部分建立全 =索引，對說明書部分建立全文索引及對申請專利範圍部 刀建立全文索引（步驟S22)。所述建立全文索引是指夢由 掃描每一部分内容的每一個詞，對每一個詞建立—個索 引，指明該詞在每一部分出現的次數及位置。 11 200915107 如圖3所示是本發明建立完索引後進行全文檢索的較 佳實施方式的流程圖。檢索模組2〇根據用戶提供的檢索標 的如關鍵字，在建立索引之後的資料庫中依專利條目進^ 專歡槽檢索’檢索到與該檢索標的相關的所有專利，並 計算檢索結果的關聯值；所述檢索結果指檢索㈣所有專 利的簡要資訊’該簡要資訊包括專利標題、部分摘要内容、 申請號等，·職專利條目可以是標題、朗書摘要、說明 書内容及申請專利範圍；關聯值為檢索結果與檢索 ===^區間為G〜1)(步驟议）。顯示模組21根據關 和值排序檢索到的專利，依序顯示上述檢索出的與當 索標的相關的專利文槽的簡要資訊，從而方便閱讀、參考 (步驟S33)。#存所述關聯值及排序結果到資料庫η中 (步驟S34)。顯賴組21根翻戶的選擇 需的專利全文或預覽專利全文（步驟S35)。狀*4不所 本發明雖以較佳實施例揭露如上，然其並非用以限定 。Γ可熟悉此項技藝者’在不脫離本發明之精神和 :’虽可做更動與潤因此本發明之保護範圍當視 後附之申請專利範圍所界定者為準。 田 【圖式簡單說明】 圖1爲本發明專利全文檢索的索引建立系統較 方式的模組圖。 1只紅 圖2爲本發明翻全文檢㈣索引建立方法 鈿方式的流程圖。 枚it實 圖3爲本發明建立完索引後進行全文檢索的較佳實施 12 200915107 方式的流程圖。 【主要元件符號說明】 伺服器 1 啟動模組 11 轉檔模組 12 索引建立模組 13 客戶端 2 檢索模組 20 顯示模組 21 網路 3 13Part of the article _ difficult, ...:! : The composition of each document, the summary number of the column, etc., the associated value is 200915107. The title includes the title of the patent document, the abstract of the manual, the content of the manual and the scope of the patent application. The retrieval method performed by the retrieval module 20 can be a single item retrieval and a multi-entry joint retrieval. The display module 21 is configured to sort the search results according to the associated values, and sequentially display the brief information of the patent documents related to the current search target retrieved in order to facilitate reading and reference. The display module 21 is further configured to download or open the patent document for the user to read when the user clicks on the retrieved patent title. The display module 21 is also used to preview the full text of the patent before downloading the full text of the patent. As shown in FIG. 2, it is a flowchart of an embodiment of the car parent of the index establishing method of the full-text search of the present invention. First, when the startup module n adds a new patent document to the detection database, the startup conversion module 12 acquires the patent document from the database 17 by the FTp (step S2). The indexing module 12 reads the contents of each component of the patent building, and saves the contents of each component of the read patent building in a predefined document format, the document of the predefined document format belongs to The type may be a web page, XML, or other type that can be converted into a text file (step S21). The index creation module 13 adds the contents of the patent documents after the above-mentioned translation to the database 17 using the index creation technique, and separately separates the contents of the respective components of the patent documents after the conversion in the database 17 Re-establishing the full-text index, including establishing a full-text index on the title, establishing a full=index in the abstract part of the specification, establishing a full-text index on the specification part, and establishing a full-text index on the patent application scope knife (step S22). The establishment of the full-text index refers to the dream of scanning each word of each part of the content, establishing an index for each word, indicating the number and location of the word in each part. 11 200915107 is a flow chart showing a preferred embodiment of the full text search after the index is established in the present invention. The search module 2 searches for all patents related to the search target according to the search target provided by the user, such as a keyword, in the database after the index is searched, according to the patent entry, and calculates the association of the search result. The search result refers to the search (4) brief information of all patents. 'The brief information includes the patent title, part of the abstract content, the application number, etc. The title of the patent can be the title, the summary of the book, the content of the manual and the scope of the patent application; The value is the search result and the search ===^ interval is G~1) (step negotiation). The display module 21 sorts the retrieved patents according to the value and the value, and sequentially displays the retrieved brief information of the patent slot associated with the target, thereby facilitating reading and reference (step S33). # Store the associated value and the sort result in the database η (step S34). It is obvious that the selection of 21 copies of the group requires the full text of the patent or the full text of the preview (step S35). The present invention is disclosed above in the preferred embodiment, but it is not intended to be limiting. 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 [Simplified illustration of the drawings] Fig. 1 is a block diagram showing a comparison of the index establishing system of the full-text search of the invention. 1 red Figure 2 is a flow chart of the method for establishing the index of the full-text (four) index of the present invention. Figure 3 is a flow chart of a preferred implementation of the full-text search after the index is established in the present invention. [Main component symbol description] Server 1 Startup module 11 Transition module 12 Index creation module 13 Client 2 Search module 20 Display module 21 Network 3 13