JPH04106663A - Patent document classifying device - Google Patents
Patent document classifying deviceInfo
- Publication number
- JPH04106663A JPH04106663A JP2225588A JP22558890A JPH04106663A JP H04106663 A JPH04106663 A JP H04106663A JP 2225588 A JP2225588 A JP 2225588A JP 22558890 A JP22558890 A JP 22558890A JP H04106663 A JPH04106663 A JP H04106663A
- Authority
- JP
- Japan
- Prior art keywords
- patent document
- technical term
- specialized field
- words
- dictionary
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 claims description 10
- 238000001514 detection method Methods 0.000 claims description 9
- 238000010586 diagram Methods 0.000 description 9
- 238000000605 extraction Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 1
Landscapes
- Machine Translation (AREA)
- Document Processing Apparatus (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
Description
【発明の詳細な説明】
（産業上の利用分野）
本発明は、特許調査作業で、目的とする専門分野に関連
する特許文書を抽出するための特許文書分類装置に関す
るものである。DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to a patent document classification device for extracting patent documents related to a target specialized field during patent search work.
（従来の技術）
特許の調査作業は、膨大な特許の資料の中から調査の対
象となる専門分野に関連する特許を抽出するもので、一
般的に、次の２種類の方法で行われている。(Prior art) Patent search work involves extracting patents related to the specialized field to be searched from a huge amount of patent data, and is generally carried out using the following two methods. There is.
第１の方法は、特許調査の対象となる専門分野に関連す
るキーワードによって、検索する方法で、たとえば、機
械翻訳機に関する特許のときは、「機械翻訳」、「構文
解析」、「訳文生成」等のキーワードを用いてこれらの
キーワードを含む特許を抽出する。また、第２の方法は
、特許の国際分類コードを用いて検索する方法で、たと
えば、機械翻訳機に関する特許の国際分類コードは、Ｇ
Ｏ６Ｆ’ｌ ５／３８、ＧＯ６Ｆ１５／２０等なので、
これらの国際分類コードが付与された特許を抽出する。The first method is to search using keywords related to the specialized field that is the subject of patent search. For example, in the case of patents related to machine translation machines, "machine translation", "syntax analysis", "translation generation" Using keywords such as , patents containing these keywords are extracted. The second method is to search using the international classification code of patents. For example, the international classification code of patents related to machine translators is
O6F'l 5/38, GO6F15/20 etc.
Extract patents assigned these international classification codes.
（発明が解決しようとする課題）
しかしながら、上記の検索方法では、キーワード又は国
際分類コードによる検索の結果得られた特許の中に、特
許調査の対象とする専門分野と無関係の特許が数多く含
まれるため、抽出された特許の中からさらに、手作業に
よって選別する必要があり、作業効率が悪いという問題
があった。(Problem to be solved by the invention) However, with the above search method, the patents obtained as a result of the search using keywords or international classification codes include many patents that are unrelated to the specialized field targeted by the patent search. Therefore, it is necessary to manually select the extracted patents, which poses a problem of poor work efficiency.
本発明は上記の問題を解決するもので、自動的に関連す
る特許文書を検索抽出できる特許文書分類装置を提供す
るものである。The present invention solves the above problems and provides a patent document classification device that can automatically search and extract related patent documents.
（課題を解決するための手段）
上記の課題を解決するため本発明は、文字の入力並びに
特許文書の分類作業に付随する指示を行う入力手段と、
特許文書の分類処理の実行を制御する実行制御部と、複
数個の特許文書を記憶する外部記憶手段と、外部記憶手
段より読み込んだ特許文書を記憶する特許文書記憶部と
、特許文書の中から専門用語と考えられる単語を検出す
る専門用語候補検出部と、専門用語候補検出部が専門用
語候補を検出するために参照する一般的な単語を格納し
た基本単語辞書と、上記専門用語候補検出部が出力する
専門用語候補を記憶する専門用語候補記憶部と、専門用
語候補記憶部に記憶された専門用語候補を基に、特許文
書がどの専門分野に属するかを判定する専門分野判定部
と、専門分野判定部が特許文書の専門分野を判定するた
めに参照する専門分野ごとにそれぞれの専門用語を格納
した少なくとも１個以上の専門用語辞書と、上記専門分
野判定部が出力する専門分野情報を格納する専門分野情
報記憶部と、特許文書の内容や特許文書の専門分野情報
を表示する表示手段とを設けるものである。(Means for Solving the Problems) In order to solve the above problems, the present invention provides an input means for inputting characters and giving instructions accompanying the classification work of patent documents;
an execution control unit that controls execution of a patent document classification process; an external storage unit that stores a plurality of patent documents; a patent document storage unit that stores patent documents read from the external storage unit; A technical term candidate detection unit that detects words that are considered to be technical terms; a basic word dictionary that stores general words that the technical term candidate detection unit refers to in order to detect technical term candidates; and the technical term candidate detection unit. a technical term candidate storage unit that stores technical term candidates output by the technical term candidate storage unit; a specialized field determination unit that determines to which specialized field the patent document belongs based on the technical term candidates stored in the technical term candidate storage unit; At least one or more technical term dictionaries that store respective technical terms for each specialized field that the specialized field determining section refers to in order to determine the specialized field of a patent document, and specialized field information output by the specialized field determining section. The invention is provided with a specialized field information storage section for storing the information, and a display means for displaying the content of the patent document and the specialized field information of the patent document.
（作 用）
上記の構成により、特許文書の専門分野を自動的に判定
して表示することができるので、従来の人手による抽出
作業がなくなり、調査の対象となる専門分野に関連する
特許文書の抽出作業の効率が大幅に改善される。(Function) With the above configuration, the specialized field of a patent document can be automatically determined and displayed, eliminating the conventional manual extraction work and allowing users to easily identify patent documents related to the specialized field that is the subject of the search. The efficiency of extraction work is greatly improved.
（実施例）
本発明の一実施例を第１図ないし第７図により説明する
。(Example) An example of the present invention will be described with reference to FIGS. 1 to 7.
第１図は、本発明による特許文書分類装置のブロック図
で、この特許文書分類装置は、文字の入力並びに特許文
書の分類作業に付随する指示を行う入力手段１と、特許
文書の分類処理の実行を制御する実行制御部２と、複数
個の特許文書を記憶する外部記憶手段３と、外部記憶手
段３より読み込んだ特許文書を記憶する特許文書記憶部
４と、特許文書の中から専門用語と考えられる単語を検
出する専門用語候補検出部５と、専門用語候補検出部５
が専門用語候補を検出するために参照する一般的な単語
を格納した基本単語辞書６と、上記の専門用語候補検出
部５が出力する専門用語候補を記憶する専門用語候補記
憶部７と、上記の専門用語候補記憶部７に記憶された専
門用語候補を基に、特許文書がどの専門分野に属するか
を判定する専門分野判定部８と、上記の専門分野判定部
８が特許文書の専門分野を判定するために参照する専門
分野ごとにそれぞれの専門用語を格納した少なくとも１
個以上の専門用語辞書９と、上記の専門分野判定部８が
出力する専門分野情報を格納する専門分野情報記憶部１
０と、特許文書の内容や特許文書の専門分野情報を表示
する表示手段１１とから構成されている。FIG. 1 is a block diagram of a patent document classification device according to the present invention. An execution control unit 2 that controls execution; an external storage unit 3 that stores a plurality of patent documents; a patent document storage unit 4 that stores patent documents read from the external storage unit 3; A technical term candidate detection unit 5 that detects words that are considered to be
a basic word dictionary 6 that stores general words to be referred to in order to detect technical term candidates; a technical term candidate storage unit 7 that stores technical term candidates output from the technical term candidate detection unit 5; A specialized field determining section 8 determines to which specialized field the patent document belongs based on the technical term candidates stored in the technical term candidate storage section 7, and the specialized field determining section 8 determines the specialized field of the patent document. At least one store containing specialized terminology for each specialized field to be referred to in order to determine the
A specialized field information storage section 1 that stores specialized field information outputted from the specialized field determining section 8 and a technical term dictionary 9 of
0, and display means 11 for displaying the content of the patent document and information on the specialized field of the patent document.
以上のように構成された特許文書分類装置の動作を、第
２図のフローチャートに沿って説明する。The operation of the patent document classification apparatus configured as described above will be explained with reference to the flowchart shown in FIG.
まず、入力子１段１から外部記憶手段３に格納されてい
る特許文書を専門分野に分類するように指示すると、実
行制御部２は、外部記憶手段３に格納されている特許文
書の中から、１文書を選択し特許文書記憶部４にその内
容を読み込む（ステップＡ）。なお、第３図は、特許文
書記憶部４に読み込まれた特許文書の内容の一例である
。First, when an instruction is given from the input element 1 stage 1 to classify patent documents stored in the external storage means 3 into specialized fields, the execution control unit 2 selects the patent documents stored in the external storage means 3 from among the patent documents stored in the external storage means 3. , one document is selected and its contents are read into the patent document storage section 4 (step A). Note that FIG. 3 is an example of the content of a patent document read into the patent document storage unit 4.
次に、実行制御部２が、専門用語候補検出部５に対して
特許文書記憶部４に読み込んだ特許文書の中から専門用
語と考えられる単語を検出するよるように指示すると、
専門用語候補記憶部５は、特許文書記憶部４に記憶され
ている特許文書のすべての単語について、基本単語辞書
６を検索し一致する単語がないかどうかを調べ、基本単
語辞書６と一致しなかった単語を専門用語候補記憶部７
に格納する（ステップＢ）。なお、基本単語辞書６は、
専門用語辞書９の登録単語を重複しないように構成する
。第４図は、第３図に示した特許文書に対する、ステッ
プＢの処理終了時に、専門用語候補記憶部７に格納され
た専門用語候補の例を示す。Next, when the execution control unit 2 instructs the technical term candidate detection unit 5 to detect words that are considered to be technical terms from the patent document read into the patent document storage unit 4,
The technical term candidate storage unit 5 searches the basic word dictionary 6 for all the words of the patent document stored in the patent document storage unit 4 to see if there are any matching words, and searches the basic word dictionary 6 for all words that match the basic word dictionary 6. Technical term candidate storage unit 7 for missing words
(Step B). The basic word dictionary 6 is
Registered words in a technical term dictionary 9 are configured so as not to overlap. FIG. 4 shows an example of technical term candidates stored in the technical term candidate storage unit 7 at the end of the process of step B for the patent document shown in FIG.
さらに、実行制御部２が、専門分野判定部８に対して専
門用語候補記憶部７に格納されている専門用語候補を基
に専門分野を判定するよう指示すると、専門分野判定部
８は専門用語候補記憶部７に格納されているすべての単
語について専門用語辞書９を検索し、それぞれの単語が
専門用語辞書９に存在するか否かを示す情報を専門分野
情報記憶部ＩＯに書き込む（ステップＣ）。第５図は、
専門用語辞書９の一例を示すもので、専門用語辞書９は
、機械翻訳機、かな漢字変換装置、プリンタの３種類の
専門用語辞書９からなっている。また、第６図は、ステ
ップＣの終了時の、判定状態を示すもので、第４図に示
した専門用語候補記憶部７に格納された専門用語候補例
について、第５図に示した専門用語辞書９を検索した結
果を専門分野情報記憶部１０に格納した例を示す。専門
分野情報記憶部ＩＯには、専門用語候補記憶部７に格納
されたそれぞれの単語について、専門用語辞書９と一致
するれば１、一致しなければＯが書き込まれる。Further, when the execution control unit 2 instructs the specialized field determining unit 8 to determine the specialized field based on the technical term candidates stored in the technical term candidate storage unit 7, the specialized field determining unit 8 The technical term dictionary 9 is searched for all the words stored in the candidate storage unit 7, and information indicating whether each word exists in the technical term dictionary 9 is written into the specialized field information storage unit IO (step C ). Figure 5 shows
This is an example of a technical term dictionary 9, which consists of three types of technical term dictionaries 9: a machine translator, a kana-kanji converter, and a printer. Further, FIG. 6 shows the judgment state at the end of step C. Regarding the technical term candidate example stored in the technical term candidate storage unit 7 shown in FIG. An example is shown in which the results of searching the terminology dictionary 9 are stored in the specialized field information storage unit 10. For each word stored in the technical term candidate storage unit 7, 1 is written in the specialized field information storage unit IO if the word matches the technical term dictionary 9, and O is written if it does not match.
なお、専門用語辞書９が複数の専門用語辞書９からなる
場合は、それぞれの専門用語辞書９ごとの検索結果が専
門分野情報記憶部１０に書き込まれ、それぞれの最下段
にそれぞれの専門用語辞書と一致する単語の合計数が書
き込まれる。In addition, when the technical term dictionary 9 consists of a plurality of technical term dictionaries 9, the search results for each technical term dictionary 9 are written in the specialized field information storage section 10, and the respective technical term dictionaries and the respective technical term dictionaries are written in the bottom row of each. The total number of matching words is written.
次に、実行制御部２は、専門分野情報記憶部ＩＯの最下
段に書き込まれた合計数を参照し、一致単語が最も多か
った専門用語辞書９、すなわち、専門用語候補記憶部７
に格納された単語を最も多く含む専門用語辞書９を調べ
、その専門用語辞書の専門分野の該特許文書の専門分野
として、表示手段１１に表示する（ステップＤ）。ここ
で、第７図は、第３図に示した特許文書の例に対して、
表示手段１１に表示された専門分野の例を示す。第６図
に示した専門分野情報記憶部１０の内容例で判るように
、各専門用語辞書９と一致する単語の合計は、機械翻訳
機専門用語辞書が８、かな漢字変換装置専門用語辞書が
４、プリンタ専門用語辞書が１となっており、機械翻訳
機専門用語辞書の一致数が最大となるため、該特許文書
の専門分野として機械翻訳機が選択された。Next, the execution control unit 2 refers to the total number written in the bottom row of the specialized field information storage unit IO, and selects the technical term dictionary 9 that has the most matching words, that is, the technical term candidate storage unit 7.
The technical term dictionary 9 containing the largest number of words stored in the terminology dictionary is checked, and the specialized field of the technical term dictionary is displayed on the display means 11 as the specialized field of the patent document (step D). Here, FIG. 7 shows the example of the patent document shown in FIG.
An example of specialized fields displayed on the display means 11 is shown. As can be seen from the content example of the specialized field information storage unit 10 shown in FIG. 6, the total number of words that match each technical term dictionary 9 is 8 for the machine translator technical term dictionary and 4 for the kana-kanji converter technical term dictionary. , the printer technical term dictionary is 1, and the machine translator technical term dictionary has the maximum number of matches, so the machine translator was selected as the specialized field of the patent document.
最後に、実行制御部２は、外部記憶手段３に格納された
すべての特許文書に対して専門分野の分類処理を終了し
たか否かを調べ（ステップＥ）、専門分野の判定を行っ
ていない文書があればステップＡへ分岐し、そうでなけ
れば、特許文書分類処理を終了する。Finally, the execution control unit 2 checks whether the specialized field classification process has been completed for all patent documents stored in the external storage means 3 (step E), and determines whether the specialized field has been determined. If there is a document, the process branches to step A; otherwise, the patent document classification process ends.
（発明の効果）
以上説明したように、本発明によれば、特許文書の中か
ら専門用語と考えられる単語を検出し、この専門用語候
補と、１個以上の専門用語辞書を照合して専門分野を判
定することにより特許文書の専門分野を自動的に判定し
て表示することができる特許文書分類装置が可能となり
、従来人手によって行っていた、専門分野に関連する特
許文書を抽出する特許調査作業の作業効率を大幅に改善
することができる。(Effects of the Invention) As explained above, according to the present invention, words that are considered to be technical terms are detected from a patent document, and this technical term candidate is compared with one or more technical term dictionaries to provide specialized terminology. By determining the field, a patent document classification device that can automatically determine and display the specialized field of a patent document has become possible, eliminating the need for patent searches to extract patent documents related to specialized fields, which was previously done manually. Work efficiency can be significantly improved.
第１図は本発明による特許文書分類装置のブロック図、
第２図はその動作を示すフローチャート、第３図は特許
文書記憶部に格納された特許文書の例を示す図、第４図
は専門用語候補記憶部に格納された内容の例を示す図、
第５図は専門用語辞書の内容の例を示す図、第６図は専
門分野情報記憶部に格納された内容の例を示す図、第７
図は表示手段への表示例を示す図である。
１ ・・・入力手段、 ２・・・実行制御部、３ ・・
・外部記憶手段、 ４ ・・・特許文書記憶部、 ５・
・・専門用語候補検出部、 ６・・・基本単語辞書、
７ ・・・専門用語候補記憶部、 ８ ・・・専門分野
判定部、 ９ ・・専門用語辞書、１０・・・専門分野
情報記憶部、１１・・・表示手段。
特許呂願人 松下電器産業株式会社
代
理
人
星
野
恒
司
第
図
第
図
第
図
く卑門用結杖桶車詰〉
扱麓酊訳樵
か乙溝↑斐挟に１
ブリ）り
１：専門用分酵書（て−玖車語刹り
Ｏ：４門用６況罎に一攻車Ｓ吉りし
第
図
第
図FIG. 1 is a block diagram of a patent document classification device according to the present invention;
FIG. 2 is a flowchart showing the operation, FIG. 3 is a diagram showing an example of a patent document stored in the patent document storage section, and FIG. 4 is a diagram showing an example of the contents stored in the technical term candidate storage section.
FIG. 5 is a diagram showing an example of the contents of the technical term dictionary, FIG. 6 is a diagram showing an example of the contents stored in the specialized field information storage unit, and FIG.
The figure is a diagram showing an example of display on the display means. 1...Input means, 2...Execution control unit, 3...
・External storage means, 4...Patent document storage unit, 5.
... technical term candidate detection unit, 6... basic word dictionary,
7... Technical term candidate storage section, 8... Specialized field determination section, 9... Technical term dictionary, 10... Specialized field information storage section, 11... Display means. Patent applicant Koji Hoshino, agent of Matsushita Electric Industrial Co., Ltd. Test book (te-kusha language 刹りO: 4 gates for 6 situations, one attack car S good luck, diagram, diagram)
Claims (1)
行う入力手段と、特許文書の分類処理の実行を制御する
実行制御部と、複数個の特許文書を記憶する外部記憶手
段と、外部記憶手段より読み込んだ特許文書を記憶する
特許文書記憶部と、特許文書の中から専門用語と考えら
れる単語を検出する専門用語候補検出部と、専門用語候
補検出部が専門用語候補を検出するために参照する一般
的な単語を格納した基本単語辞書と、上記専門用語候補
検出部が出力する専門用語候補を記憶する専門用語候補
記憶部と、専門用語候補記憶部に記憶された専門用語候
補を基に、特許文書の専門分野を判定する専門分野判定
部と、専門分野判定部が特許文書の専門分野を判定する
ために参照する専門分野ごとにそれぞれの専門用語を格
納した少なくとも１個以上の専門用語辞書と、上記専門
分野判定部が出力する専門分野情報を格納する専門分野
情報憶部と、特許文書の内容や特許文書の専門分野情報
を表示する表示手段とを設けたことを特徴とする特許文
書分類装置。An input means for inputting characters and giving instructions related to the classification work of patent documents; an execution control section for controlling execution of the classification process of patent documents; an external storage means for storing a plurality of patent documents; and an external storage means a patent document storage unit that stores patent documents read from the patent document; a technical term candidate detection unit that detects words that are considered to be technical terms from the patent document; A basic word dictionary that stores common words to be used; a technical term candidate storage unit that stores technical term candidates output from the technical term candidate detection unit; , a specialized field determining unit that determines the specialized field of a patent document, and at least one or more technical terms storing each specialized term for each specialized field that the specialized field determining unit refers to in order to determine the specialized field of a patent document. A patent characterized in that it is provided with a dictionary, a specialized field information storage unit that stores the specialized field information outputted by the specialized field determining unit, and a display means that displays the content of the patent document and the specialized field information of the patent document. Document classification device.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2225588A JPH04106663A (en) | 1990-08-28 | 1990-08-28 | Patent document classifying device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2225588A JPH04106663A (en) | 1990-08-28 | 1990-08-28 | Patent document classifying device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| JPH04106663A true JPH04106663A (en) | 1992-04-08 |
Family
ID=16831669
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP2225588A Pending JPH04106663A (en) | 1990-08-28 | 1990-08-28 | Patent document classifying device |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JPH04106663A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH08305726A (en) * | 1995-04-28 | 1996-11-22 | Fuji Xerox Co Ltd | Information retrieving device |
| JPH09265482A (en) * | 1996-01-26 | 1997-10-07 | Mitsubishi Electric Corp | Database retrieval device and database retrieval method |
| JP2005128872A (en) * | 2003-10-24 | 2005-05-19 | Toshiba Solutions Corp | Document retrieving system and document retrieving program |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH02105973A (en) * | 1988-10-14 | 1990-04-18 | Nippon Telegr & Teleph Corp <Ntt> | Automatic classifying device for document |
-
1990
- 1990-08-28 JP JP2225588A patent/JPH04106663A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH02105973A (en) * | 1988-10-14 | 1990-04-18 | Nippon Telegr & Teleph Corp <Ntt> | Automatic classifying device for document |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH08305726A (en) * | 1995-04-28 | 1996-11-22 | Fuji Xerox Co Ltd | Information retrieving device |
| JPH09265482A (en) * | 1996-01-26 | 1997-10-07 | Mitsubishi Electric Corp | Database retrieval device and database retrieval method |
| JP2005128872A (en) * | 2003-10-24 | 2005-05-19 | Toshiba Solutions Corp | Document retrieving system and document retrieving program |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US5745745A (en) | Text search method and apparatus for structured documents | |
| US20020143537A1 (en) | Process of automatically generating translation- example dictionary, program product, computer-readable recording medium and apparatus for performing thereof | |
| KR100798752B1 (en) | Apparatus for and method of korean orthography | |
| JPH08235182A (en) | Method and device for document processing | |
| JPH04106663A (en) | Patent document classifying device | |
| JPH04106664A (en) | Patent document classifying device | |
| JP2792147B2 (en) | Character processing method and device | |
| JP3955410B2 (en) | Similar information collating device, similar information collating method, and recording medium recording similar information collating program | |
| KR102338949B1 (en) | System for Supporting Translation of Technical Sentences | |
| JPH08115330A (en) | Method for retrieving similar document and device therefor | |
| JP3388153B2 (en) | Machine translation device and computer-readable recording medium recording machine translation program | |
| JP4769379B2 (en) | Document search device | |
| JPH0635971A (en) | Document retrieving device | |
| JPH1115826A (en) | Document analyzer and its method | |
| JP2005189955A (en) | Document processing method, document processor, control program, and recording medium | |
| JPH01259448A (en) | Kanji input system | |
| JP2936761B2 (en) | Proofreading device for Japanese documents | |
| JP3054872B2 (en) | Machine translator | |
| JPH06309369A (en) | Information retrieving device | |
| JPH05334302A (en) | Method and device for document processing | |
| JPS6368989A (en) | Document reader | |
| JPH0863464A (en) | Retrieving method for information processor | |
| JPH0365763A (en) | Document retrieving device | |
| JPH05233619A (en) | Method for correcting error of japanese language sentence and device therefor | |
| JP2005189954A (en) | Document processing method, document processor, control program, and recording medium |