JPS58117039A - Kana(japanese syllabary)/kanji(chinese character) conversion system - Google Patents
Kana(japanese syllabary)/kanji(chinese character) conversion systemInfo
- Publication number
- JPS58117039A JPS58117039A JP56210676A JP21067681A JPS58117039A JP S58117039 A JPS58117039 A JP S58117039A JP 56210676 A JP56210676 A JP 56210676A JP 21067681 A JP21067681 A JP 21067681A JP S58117039 A JPS58117039 A JP S58117039A
- Authority
- JP
- Japan
- Prior art keywords
- words
- kana
- connection
- kanji
- possibility
- 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
Landscapes
- Document Processing Apparatus (AREA)
Abstract
Description
【発明の詳細な説明】
(1) 発明の技術分野
零発ll紘、文法解析機構を備え九仮虐漢字変換方式K
>いて、単語の接続性に優先度をもえせえものである・
・
■ 従来技術と問題点
第1図は仮名漢字変換方式の八−ドク、ア構威の!実施
例を示す亀のであって、lはキーボード、2は中央処理
装置、3はディスプレイ、4は補助記憶装置、5はメモ
リをそれぞれ示している。キーボード1は仮名文字列を
入力するためのものであ〉、中央処理装置2は仮名漢字
変換の制御を行うものである。ディスプレイ3は変換結
果を表示するものである。補助記憶装置4の中には単一
辞書が格納され、この単一辞書の中には仮名表記の単語
、漢字表記の単語、単lIIの文法情報および早躇の出
現頻度が記入されている。メモリ器の中には、品詞と品
詞との間の接続の可能性を示す接続テーブルが格納され
ている。[Detailed description of the invention] (1) Technical field of the invention
>In other words, priority is given to the connectivity of words.
・ ■ Conventional technology and problems Figure 1 shows the 8-doku, A-structure of the Kana-Kanji conversion method! In the figure showing the embodiment, 1 is a keyboard, 2 is a central processing unit, 3 is a display, 4 is an auxiliary storage device, and 5 is a memory. The keyboard 1 is for inputting kana character strings, and the central processing unit 2 is for controlling kana-kanji conversion. The display 3 is for displaying the conversion results. A single dictionary is stored in the auxiliary storage device 4, and words written in kana, words written in kanji, grammatical information in single letters, and frequency of occurrence are entered in this single dictionary. A connection table indicating possible connections between parts of speech is stored in the memory device.
従来の仮鳴漢字変換は一般に下記のようにして行われて
いる。仮名文字列もしくはローマ字列でベタ書きされた
複数文節を漢字に変換する様、まず、単、語の列に分制
す石、こ−で、分割された単語が接続可能か否かを、各
単語の文法情報と接続テーブルとを用いて判定し、接続
可能な単、語だけを候補として残す、そして、残りた候
補の中から長さおよびS度が最大のものを仮名漢字変換
の結果とする。Conventional kanai-kanji conversion is generally performed as follows. In order to convert multiple phrases written solidly in kana or romaji strings into kanji, we first use a stone that separates each word into a string of single words to determine whether or not the divided words can be connected. This is determined using the grammatical information and connection table, and only connectable words are left as candidates, and among the remaining candidates, the one with the largest length and S degree is used as the result of kana-kanji conversion. .
第2図は従来の接続テーブルの構成を示すものである。FIG. 2 shows the structure of a conventional connection table.
接続テーブル6は品詞と品詞との間の接続の可能性を示
す亀のであつて、接続の可能性のある品詞間の交点には
「1」が書込まれ、接続の可能性のない品詞間の交点に
はrOJが書込まれている。第2図の接続テーブルにお
い′C祉、地名と数詞との間には接続の可能性があるの
で、これらの交点は「1」とされ、地名と地名接尾語と
の間にも接続の可能性があるので、これらの支点も「1
」とされている、地名と数詞との間に接続の可能性のあ
ること社、「東京23区」、「東北6県」という表現の
例からみて明らかであゐ。The connection table 6 is a turtle that shows the possibility of connection between parts of speech. ``1'' is written at the intersection between parts of speech that have a possibility of connection, and between parts of speech that have no possibility of connection. rOJ is written at the intersection. In the connection table in Figure 2, there is a possibility of a connection between a place name and a number word, so the intersection of these is set to ``1'', and a connection is also possible between a place name and a place name suffix. These fulcrums also have “1”
It is clear from the examples of expressions such as ``Tokyo's 23 wards'' and ``Tohoku 6 prefectures'' that there is a possibility of a connection between a place name and a number word.
第28のような接続テーブルを用いる従来方式では、評
価関数として
fに)=Xの頻度×32
を用いたとすると1文法的に接続可能であり光場合には
、あとは単語の長さとS変だけによって仮名漢字変換の
結果が決定されてしまう、この際。In the conventional method using a connection table such as No. 28, if f is used as the evaluation function = frequency of In this case, the result of kana-kanji conversion is determined only by
単語と単語の1接続の可能性の大きさ0というものへの
考慮が一切払われていない、ヒのため、「四」の出現頻
度がr市」の出JIIs度よ〉大きいとすると、例えば
「璽プ^マV」という仮名文字列が「横浜市」と変換さ
れるべきところを「横浜1 四」と変換されてしまう、
「横浜」は地鳴であ勤、「四」は長さ1の数詞であシ、
r市」杜長さ1の地名接尾語である。Since there is no consideration given to the fact that the probability of a word-to-word connection is 0, if the frequency of appearance of ``shi'' is greater than the occurrence of ``r city'', for example, The kana character string "Seal Puma V" should be converted to "Yokohama City" but instead is converted to "Yokohama 1 4".
``Yokohama'' is an earth sound, and ``four'' is a number word with a length of 1.
"r city" is a place name suffix with a length of 1.
の) 発明の目的
本発明は、上記の考111K基づく4のであjl、単語
間の接続の可能性を考慮すゐことによりて仮名漢字変換
皐を向上させ得るようにした仮名漢字変換方式を提供す
ゐことを目的としている。Purpose of the Invention The present invention is based on the above-mentioned idea 111K, and provides a kana-kanji conversion method that can improve kana-kanji conversion by taking into account the possibility of connections between words. The purpose is to be sui.
(4) 発明の構成
そし【そのため、本発明の仮名漢字変換方式は、仮名表
記の単語、漢字表□記の単語、単語の文法情報およ・び
単語の出現頻度が記入された単語辞書、並びに単語間の
接続情報が記入され九接続テーブルを備え、上記単語辞
書および接続テーブルを用いて入力された仮置゛文字列
を漢字混り文に変換する仮名漢字変換方式において、上
記接続テーブル・のΦに単語間の接続の可能性の大きさ
を示すn進数(たりし、nは意よ妙大)の数値を書込み
、蟲該接続の可能性の大きさを示す数値を単語抽出の際
のパラメータの1つとしたことを**とするものである
。 ゛
(2)”□発明の実施例 ゛
′以下、本奥明を回置を参照しつつ説明する。(4) Structure of the invention [Therefore, the kana-kanji conversion method of the present invention uses a word dictionary in which words written in kana, words written in kanji, grammatical information of the words, and frequency of occurrence of the words are entered; In the kana-kanji conversion method, which has nine connection tables in which connection information between words is entered, and converts a temporary character string inputted using the word dictionary and the connection table into a sentence containing kanji, the connection table and When extracting words, write an n-ary number (Tarishi, n is meaning) in Φ that indicates the possibility of a connection between words. ** indicates that it is one of the parameters.゛(2)”□Embodiments of the invention ゛
'Hereafter, I will explain the present Okumei with reference to inversion.
本発明は、単語と単語との接続が可能であるか否かのチ
ェックをする九めに使用されていた接続テーブルに1接
続の可能性の大きさ1という概念を導入したものである
。第3′図は本発明の接続テーブルを説明すbものであ
って、6パは接続テーブルを示している6本発明の接続
テーブル6’Kkいては、品詞と品詞との間の交点II
I続の可能性の大きさを示す数値が記入されてい、る、
数値が大きい程、接続の可能性が大きいζどを示してい
る。The present invention introduces the concept of the probability of one connection being 1 into the connection table that was used to check whether or not words can be connected. FIG. 3' is a diagram illustrating the connection table of the present invention, where 6 pa indicates the connection table 6'Kk, the intersection point II between the parts of speech and the
A numerical value indicating the size of the possibility of continuation is entered.
The larger the value, the greater the possibility of connection.
図示の例では、地名と数詞との交点には数値1!1が記
入され、地名と地名接尾語との交・点には数値12″が
記入されている。勿・論、縦方向の1品詞社前の品詞で
あ夛、横方向の、品詞は辷れに続く品詞である。評価関
数の決め方は・0自であ、るが、例えば。In the illustrated example, the number 1!1 is written at the intersection between the place name and the numeral, and the number 12'' is written at the intersection/point between the place name and the place name suffix.Of course, 1 in the vertical direction The part of speech in front of the part of speech is the part of speech in the front, and the part of speech in the horizontal direction is the part of speech that follows the part of speech.The way to determine the evaluation function is 0.
fに)=1のS度×3・2×(前半−の接続度)とする
仁とが出来る。前例と同じように、「冒プハマν」を仮
名漢字変換する場合を考えると、。f) = 1 S degree x 3.2 x (first half - connection degree). As in the previous example, let's consider converting ``Bakuhama ν'' into kana and kanji.
「四」の評価関数f(四)は、
f(四):=:1ooOX32’X1
とな夛、/(市)の評価関数は、
f(市)=900X3!’X意
となる。/(&!g)</(市)であるので、「11:
yへマV」は「横浜市」と変換される。なか、単一(四
)の出現頻度を”1000@、単−r市」の出現頻度を
1900”と仮定している。The evaluation function f(four) for "four" is f(four):=:1ooOX32'X1, and the evaluation function for /(city) is f(city)=900X3! 'It becomes X intention. /(&!g)</(city), so "11:
yhemaV" is converted to "Yokohama City." It is assumed that the appearance frequency of single (4) is "1000@, and the appearance frequency of single-r city" is 1900.
(0発明の効果
以上の説明から明らかなように、本発明によれば、一度
に入力され九複数文節を高い変換率で以て漢字に変換す
ることが出来る。(0) Effects of the Invention As is clear from the above description, according to the present invention, nine or more phrases input at one time can be converted into kanji with a high conversion rate.
第1図は仮名漢字変換装置のハードフェア構成の1例を
示す図、第2図は従来の接続テーブルを説明する図、第
3図は本発明の接続テーブルを説明する図である。
1・・・キーボード、!・・・中央処理装置、3・・・
ディスプレイ、5・・・メモリ、6と6′・・・接続テ
ーブル。
特許出願人富士通株式会社
代珈人弁珊士 京 谷 四 部FIG. 1 is a diagram illustrating an example of the hardware configuration of a kana-kanji conversion device, FIG. 2 is a diagram illustrating a conventional connection table, and FIG. 3 is a diagram illustrating a connection table according to the present invention. 1...Keyboard! ...Central processing unit, 3...
Display, 5...Memory, 6 and 6'...Connection table. Patent Applicant: Fujitsu Ltd. Attorney: Yobu Kyotani
Claims (1)
び単語の出現頻度が記入され九単語辞書、並びに単語間
の接続情報が記入され光接続テーブルを備え、上記単一
辞書および接続テーブルを用いて入力され光仮名文字列
を漢・字混シ文に変換する仮名漢字変換方式において、
上記接続テーブルの中に単語間の接続の可能性の大きさ
を示すn進数(たゾし、nは2よp大)の数値を書込み
、轟該接続の可能性の大きさを示す数値を単一抽出の際
のパラメータの!つとしたことを特徴とする仮名漢字変
換方式。It is equipped with a nine-word dictionary in which words written in kana, words written in kanji, grammatical information of words, and frequency of appearance of words are entered, and an optical connection table in which connection information between words is entered, and using the above single dictionary and connection table. In the kana-kanji conversion method, which converts the light kana character string input by
In the connection table above, write an n-adic number (n is greater than 2) that indicates the possibility of a connection between words, and write a number that indicates the possibility of a connection between words. of parameters during single extraction! A Kana-Kanji conversion method that is characterized by:
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP56210676A JPS58117039A (en) | 1981-12-30 | 1981-12-30 | Kana(japanese syllabary)/kanji(chinese character) conversion system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP56210676A JPS58117039A (en) | 1981-12-30 | 1981-12-30 | Kana(japanese syllabary)/kanji(chinese character) conversion system |
Publications (1)
Publication Number | Publication Date |
---|---|
JPS58117039A true JPS58117039A (en) | 1983-07-12 |
Family
ID=16593261
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
JP56210676A Pending JPS58117039A (en) | 1981-12-30 | 1981-12-30 | Kana(japanese syllabary)/kanji(chinese character) conversion system |
Country Status (1)
Country | Link |
---|---|
JP (1) | JPS58117039A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5323310A (en) * | 1991-02-14 | 1994-06-21 | The British And Foreign Bible Society | Analyzing textual documents |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS5638665A (en) * | 1980-04-02 | 1981-04-13 | Canon Inc | Character processor |
-
1981
- 1981-12-30 JP JP56210676A patent/JPS58117039A/en active Pending
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS5638665A (en) * | 1980-04-02 | 1981-04-13 | Canon Inc | Character processor |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5323310A (en) * | 1991-02-14 | 1994-06-21 | The British And Foreign Bible Society | Analyzing textual documents |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US4653100A (en) | Audio response terminal for use with data processing systems | |
JPS58117039A (en) | Kana(japanese syllabary)/kanji(chinese character) conversion system | |
JPS60247770A (en) | Character processor | |
JPS6037510B2 (en) | Romaji-kanji conversion method | |
JPS6282457A (en) | Method and device for inputting word of foreign origin word in japanese word processor | |
JPS628259A (en) | Kana (japanese syllabary)-kanji (chinese character) converting system | |
JPH0441399Y2 (en) | ||
JPS61233862A (en) | Kana-kanji converter | |
KR960001103B1 (en) | Computer i/o system using ro-otza | |
JPS5832418B2 (en) | Kanji-mixed sentence input device | |
JPS6132171A (en) | Document forming device having adjunct key group | |
JPS59121425A (en) | Chinese phonetic alphabet of kanji converter | |
JPS62118469A (en) | Loan word conversion system for word processor | |
JP2744241B2 (en) | Character processor | |
JPS6246030B2 (en) | ||
CN1010620B (en) | Chinese character processing device | |
JPS622350A (en) | Kana to kanji conversion system | |
JPH03222060A (en) | Japanese reading system | |
JPS60217452A (en) | Japanese word input processing unit | |
DIL | Review 39--No Title | |
JPS6327966A (en) | Katakana conversion system in word processor | |
JPH04127363A (en) | Note sound/alphabetic character combination code control system | |
JPS59168533A (en) | Converting system of character string | |
JPH01116850U (en) | ||
JPS6175471A (en) | Logotype learning system |