CA1251570A - Bilingual translation system with self intelligence - Google Patents
Bilingual translation system with self intelligenceInfo
- Publication number
- CA1251570A CA1251570A CA000509020A CA509020A CA1251570A CA 1251570 A CA1251570 A CA 1251570A CA 000509020 A CA000509020 A CA 000509020A CA 509020 A CA509020 A CA 509020A CA 1251570 A CA1251570 A CA 1251570A
- Authority
- CA
- Canada
- Prior art keywords
- word
- words
- translation
- language
- learned
- 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.)
- Expired
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/211—Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/268—Morphological analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/40—Processing or translation of natural language
- G06F40/42—Data-driven translation
- G06F40/47—Machine-assisted translation, e.g. using translation memory
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/40—Processing or translation of natural language
- G06F40/55—Rule-based translation
Abstract
Abstract A translation system for performing a translation from a first language into a second language in an interaction mode between the system and an operator, comprising means for storing words of the second language translated from source words used in a source sentence of the first lan-guage. One of the stored words is selected as a learned word at the time of performing a translation and is dis-played. The machine converts the sentence of the first language into a sentence of the second language using the selected learned word. The system enables improvements in the correctness and the speed of the translation.
Description
;7~3 Bilingual translation system with self intelligence The present invention relates to an automatic trans-lation system between two languages, with self learning ability or self intelligible ability.
To enable the prior art to be described with the aid of diagrams, the figures of the drawings will first be listed.
Fig. 1 is a block diagram showing a process of bi-lingual translation, Fig. 2 is a block diagram showing an embodiment of a translation system according to the present invention, Fig. 3 is a block diagram showing an example of a translation module used in the system of Fig. 2, Fig. 4 is a block diagram showing a further detail of the translation module, Figs. 5(a), 5(b) and 5(c) are schematic diagrams showing a feature of storing the input language in a buffer in the translation system shown in Fig. 2, Figs. 6(a) and 6(b) are schematic diagrams showing grammatical trees used in the translation system shown i~ Fig. 2, Fig. 7 is a flow chart showing operation of an embodiment of the translation system according to the present invention, Fig. 8 is a flow chart showing a modification of the operation of the translation machine according to 1'~51~
To enable the prior art to be described with the aid of diagrams, the figures of the drawings will first be listed.
Fig. 1 is a block diagram showing a process of bi-lingual translation, Fig. 2 is a block diagram showing an embodiment of a translation system according to the present invention, Fig. 3 is a block diagram showing an example of a translation module used in the system of Fig. 2, Fig. 4 is a block diagram showing a further detail of the translation module, Figs. 5(a), 5(b) and 5(c) are schematic diagrams showing a feature of storing the input language in a buffer in the translation system shown in Fig. 2, Figs. 6(a) and 6(b) are schematic diagrams showing grammatical trees used in the translation system shown i~ Fig. 2, Fig. 7 is a flow chart showing operation of an embodiment of the translation system according to the present invention, Fig. 8 is a flow chart showing a modification of the operation of the translation machine according to 1'~51~
2 --the present invention, Fig. 9 is a block diagram showing another embodiment of translation sys~em according to the present invention, and Figs. lO(a) and lO(b) are flow charts showing essen-tial portions of the operation of the translation system shown in Fig. 9.
In general, bilingual translation using a transla-tion machine is performed in the manner shown in Fig. 1, wherein the originating sentence of the source language to be translated must be analyzed in various ways in the translation machine. These analyses can be classified into morpheme analysis, sente~ce construction analysis or syntax analysis, and meaning analysis. The morpheme analysis is to classify each of the words into the per-son, number and sequence of the sentence, by reference to grammatical information, translation information and parts of speech from a dictionary contained in the trans-lation machine. The syntax analysis is to analyze the construction of the sentence by checking the grammatical relationship between the words. The meaning analysis is to determine a correct analysis on the basis of a plura-lity of the syntax analyses. The machine translation is made by performing the morpheme analysis, syntax analy-sis and meaning analysis up to a predetermined level, to obtain an internal construction of the sentence of the original language, and thereafter the trans'ation machine converts the inner construction of the original language into an internal construction of the sentence of the trans~ation language corresponding to the predetermined level. Then the translation machine generates an output translation in the desired language. The accuracy of the translation depends on the height of the predetermined level of analysis. A translation made by using only the morpheme analysis cannot realize a translation of the ~f~S~
sentence basis, and such translation is limited to a word basis translation, as performed in a handy type of electronic translator. A translation machine perform-ing morpheme analysis and syntax analysis can translate with grammatical correctness, but generates a plurality of translation results, so that the operator must select the correct translation among them, thus increasing the work of the operator. A translation machine performing up to meaning analysis is theoretically able to output only one correct translation result, but a great deal of information ~ust be provided in the translation machine. Manufacture of such translation machine therefore becomes very difficult if not impossible.
A way of translation in a conventional translation machine is explained hereinafter with reference to examples of sentences such as I write a letter And, I mail the letter.
It is assumed that the dictionary of the translation machine stores the translated words of the word ~'letter"
as "moji ~ (a character)" first and subsequently as ~tegami q~ (a letter) n in Japanese. In this example, the translation machine generates the translation of the above sentence in Japanese that (watakushi wa moji o kaku). The user may converse with the translation machine to select a desired translation of ~L~ (watakushi wa tegam~ o kaku).
The conventional translation machine without any self intelligence ability generates a translation of the second sentence as ~L~ ~ (watakushi wa ~ o yu~o suru) as a primary translation.
Accordingly, the operator must change the word "moji"
into "tegami" again.
An essential object of the present invention is to provide a translation system that is able to generate a ~ZS15 ~
correct translation with a high degree of correctness.
Another object of the present invention is to provide a translation system that is able to generate a correct translation at high speed.
~ further object of the present invention is to provide a translation system that is able to generate a correct translation by means of an easy operation.
A still further object of the present invention ;s to provide a translation system that is able to perform translation using learned words with the ~irst priority so as to proceed with the translation quickly.
According to the present invention, there is provided a translation system for translating sentences in a first language to sentences in a second language, comprising:
storage means for storing individual words in said first language together with a plurality of words in said second language, each word in said plurality of words being equivalent to the individual word stored with said plur-ality of words; means for storing an input sentence in said first language to be translated; means for determining the part of speech of each individual word in said stored input sentence; means for selecting a first one of said plurality of words in said second language stored with an individual word as corresponding to said individual word based on the part of speech determination; means for selecting another one of said plurality of words as corres-ponding to said individual word in response to a signal from a user indicating the previous selected word to be incorrect; means for storing a selected word in said second language as a learned word corresponding to an individual word in said first language in response to a signal from a user indicating the correspondence to be correct; and means for selecting the learned word as said first one of said plurality of words in subsequent translations of input sentences containing said individual word.
7~
- 4a -Referring to Fig. 2, an example of a translation machine according to the present invention comprises a CPU (central processing unit) 1 consisting of a micro~
computer provided with a translation program for trans-lating English into Japanese, a main memory 2, a display unit CRT 3, a key board 4 for inputting various data Eor translation, such as alphabetical characters, numerals and Japanese characters. The machine performs the trans]ation using an interaction system between the machine and the operator. A translation module 5 is coupled to the CPU 1, main memory ~, display unit 3 and key board 4 thro~gh a bus 10. The translati.on module 5 is also 1~51~7~
coupled with an English to Japanese dictionary memory 6 storing grammatical rules of English and Japanese and a tree construction conversion rule for translation of English into Japanese.
The key board 4 includes a start key for starting the translation machine, ten keys for inputting numer-ical characters of 0 to 9, character keys and various keys for performing English to Japanese translation as well as a self intelligence key. Moreover, there are provided in the key board a learning mode set key, and an automatic learning key (not shown).
In the translation machine shown in Fig. 2, the source sentence of the source langua~e inputted by the key board 4 is transmitted to the translation module 5 under the control of the CPU 1 and the result of the translation is transmitted to the unit 3 for display. The translation module 5 comprises a plurality of buffers A to F control-led by the CPU 1 following the program shown in Fig. 7.
When a source English sentence ~This is a pen" is inputted from the key board 4, the source sentence is stored in the buffer A as shown in Fig. 5(a). Neces-sary information is consulted in the dictionary part 11 (Fig. 3) in the translation module 5, and the selected information is stored in the buffer B. Part of speech information for each of the words of the source sentence thus drawn from the dictionary part 11 is stored in the buffer B as shown in Fig. 5(b). The word "this" is de-finitely selected by the morpheme analysis unit 12 and the construction relation of each of the words is stored in the buffer ~ as shown in Fig. 6(a). Using the grammar rule stored in the grammar memory 6 shown in Fig. 2, the following information can be obtained.
a sentence a sub~ect part and a predicate part the subject part noun phrase 35 predicate part verb and noun phrase noun phrase a pronoun noun phrase article and noun The above results mean that a sentence comprises a sub-ject part and a predicate part, for example.
In a conversion unit 13, a syntax analysis is per-formed according to the sentence construction tree and is stored in the buffer D, as shown in Fig. 6(b). The result stored in the buffer D is added by one or more suitable auxiliary verbs in the translation generation unit 14 so as to provide a suitable sentence translated into Japanese, which latter is stored in the buffer E.
The translation in Japanese is then outputted from the translation generation unit 14.
A self learning buffer 2a is provided in the main memory 2.
As shown in Fig. 7(a), when the translation machine is started, the self learning buffer 2a (represented by S L B) is cleared in step S 1. The operator selects the self learning mode (represented by S L M) by operation of the self learning key (not shown). This operation is detected in step S 2. If the self learning mode is not set, the program flow goes to step S 8 wherein the non self learning mode (represented by N S L M) is set. If the self learning mode is set, the program flow goes to step S 3 wherein it is judged whether or not an initial self learning mode (I S L M) is made. The initial self learning mode is a mode of storing predetermined know-ledge of translation in the machine. For example, if the machine initially learns the word "field", upon in-put of the word "field", the Japanese words for "field"
are displayed on the display unit 3 as in Table 1.
TABL~ 1 source word "field"
Japanese 1. nohara 2. hirogari ~25157~
In general, bilingual translation using a transla-tion machine is performed in the manner shown in Fig. 1, wherein the originating sentence of the source language to be translated must be analyzed in various ways in the translation machine. These analyses can be classified into morpheme analysis, sente~ce construction analysis or syntax analysis, and meaning analysis. The morpheme analysis is to classify each of the words into the per-son, number and sequence of the sentence, by reference to grammatical information, translation information and parts of speech from a dictionary contained in the trans-lation machine. The syntax analysis is to analyze the construction of the sentence by checking the grammatical relationship between the words. The meaning analysis is to determine a correct analysis on the basis of a plura-lity of the syntax analyses. The machine translation is made by performing the morpheme analysis, syntax analy-sis and meaning analysis up to a predetermined level, to obtain an internal construction of the sentence of the original language, and thereafter the trans'ation machine converts the inner construction of the original language into an internal construction of the sentence of the trans~ation language corresponding to the predetermined level. Then the translation machine generates an output translation in the desired language. The accuracy of the translation depends on the height of the predetermined level of analysis. A translation made by using only the morpheme analysis cannot realize a translation of the ~f~S~
sentence basis, and such translation is limited to a word basis translation, as performed in a handy type of electronic translator. A translation machine perform-ing morpheme analysis and syntax analysis can translate with grammatical correctness, but generates a plurality of translation results, so that the operator must select the correct translation among them, thus increasing the work of the operator. A translation machine performing up to meaning analysis is theoretically able to output only one correct translation result, but a great deal of information ~ust be provided in the translation machine. Manufacture of such translation machine therefore becomes very difficult if not impossible.
A way of translation in a conventional translation machine is explained hereinafter with reference to examples of sentences such as I write a letter And, I mail the letter.
It is assumed that the dictionary of the translation machine stores the translated words of the word ~'letter"
as "moji ~ (a character)" first and subsequently as ~tegami q~ (a letter) n in Japanese. In this example, the translation machine generates the translation of the above sentence in Japanese that (watakushi wa moji o kaku). The user may converse with the translation machine to select a desired translation of ~L~ (watakushi wa tegam~ o kaku).
The conventional translation machine without any self intelligence ability generates a translation of the second sentence as ~L~ ~ (watakushi wa ~ o yu~o suru) as a primary translation.
Accordingly, the operator must change the word "moji"
into "tegami" again.
An essential object of the present invention is to provide a translation system that is able to generate a ~ZS15 ~
correct translation with a high degree of correctness.
Another object of the present invention is to provide a translation system that is able to generate a correct translation at high speed.
~ further object of the present invention is to provide a translation system that is able to generate a correct translation by means of an easy operation.
A still further object of the present invention ;s to provide a translation system that is able to perform translation using learned words with the ~irst priority so as to proceed with the translation quickly.
According to the present invention, there is provided a translation system for translating sentences in a first language to sentences in a second language, comprising:
storage means for storing individual words in said first language together with a plurality of words in said second language, each word in said plurality of words being equivalent to the individual word stored with said plur-ality of words; means for storing an input sentence in said first language to be translated; means for determining the part of speech of each individual word in said stored input sentence; means for selecting a first one of said plurality of words in said second language stored with an individual word as corresponding to said individual word based on the part of speech determination; means for selecting another one of said plurality of words as corres-ponding to said individual word in response to a signal from a user indicating the previous selected word to be incorrect; means for storing a selected word in said second language as a learned word corresponding to an individual word in said first language in response to a signal from a user indicating the correspondence to be correct; and means for selecting the learned word as said first one of said plurality of words in subsequent translations of input sentences containing said individual word.
7~
- 4a -Referring to Fig. 2, an example of a translation machine according to the present invention comprises a CPU (central processing unit) 1 consisting of a micro~
computer provided with a translation program for trans-lating English into Japanese, a main memory 2, a display unit CRT 3, a key board 4 for inputting various data Eor translation, such as alphabetical characters, numerals and Japanese characters. The machine performs the trans]ation using an interaction system between the machine and the operator. A translation module 5 is coupled to the CPU 1, main memory ~, display unit 3 and key board 4 thro~gh a bus 10. The translati.on module 5 is also 1~51~7~
coupled with an English to Japanese dictionary memory 6 storing grammatical rules of English and Japanese and a tree construction conversion rule for translation of English into Japanese.
The key board 4 includes a start key for starting the translation machine, ten keys for inputting numer-ical characters of 0 to 9, character keys and various keys for performing English to Japanese translation as well as a self intelligence key. Moreover, there are provided in the key board a learning mode set key, and an automatic learning key (not shown).
In the translation machine shown in Fig. 2, the source sentence of the source langua~e inputted by the key board 4 is transmitted to the translation module 5 under the control of the CPU 1 and the result of the translation is transmitted to the unit 3 for display. The translation module 5 comprises a plurality of buffers A to F control-led by the CPU 1 following the program shown in Fig. 7.
When a source English sentence ~This is a pen" is inputted from the key board 4, the source sentence is stored in the buffer A as shown in Fig. 5(a). Neces-sary information is consulted in the dictionary part 11 (Fig. 3) in the translation module 5, and the selected information is stored in the buffer B. Part of speech information for each of the words of the source sentence thus drawn from the dictionary part 11 is stored in the buffer B as shown in Fig. 5(b). The word "this" is de-finitely selected by the morpheme analysis unit 12 and the construction relation of each of the words is stored in the buffer ~ as shown in Fig. 6(a). Using the grammar rule stored in the grammar memory 6 shown in Fig. 2, the following information can be obtained.
a sentence a sub~ect part and a predicate part the subject part noun phrase 35 predicate part verb and noun phrase noun phrase a pronoun noun phrase article and noun The above results mean that a sentence comprises a sub-ject part and a predicate part, for example.
In a conversion unit 13, a syntax analysis is per-formed according to the sentence construction tree and is stored in the buffer D, as shown in Fig. 6(b). The result stored in the buffer D is added by one or more suitable auxiliary verbs in the translation generation unit 14 so as to provide a suitable sentence translated into Japanese, which latter is stored in the buffer E.
The translation in Japanese is then outputted from the translation generation unit 14.
A self learning buffer 2a is provided in the main memory 2.
As shown in Fig. 7(a), when the translation machine is started, the self learning buffer 2a (represented by S L B) is cleared in step S 1. The operator selects the self learning mode (represented by S L M) by operation of the self learning key (not shown). This operation is detected in step S 2. If the self learning mode is not set, the program flow goes to step S 8 wherein the non self learning mode (represented by N S L M) is set. If the self learning mode is set, the program flow goes to step S 3 wherein it is judged whether or not an initial self learning mode (I S L M) is made. The initial self learning mode is a mode of storing predetermined know-ledge of translation in the machine. For example, if the machine initially learns the word "field", upon in-put of the word "field", the Japanese words for "field"
are displayed on the display unit 3 as in Table 1.
TABL~ 1 source word "field"
Japanese 1. nohara 2. hirogari ~25157~
3. maizochi ~. senjo 5. kyogijo 6. jimen 7. bunya 8. ba The operator then moves the cursor below the Japanese word bunya, and "bunya" is stored in the main memory 2.
Thus, every time the word "field" appears, the Japanese word "bunya" can be selected with first priority. If the operator wishes to delet~ the Japanese words alreadystored in the main memory, the unnecessary word can be deleted from the main memory ~ by use of a clear key in step S ~ In step S 5 it is judged whether or not an automatic self learning mode A S L M is set. I~ the A S L M is set, the program flow goes to step S 6 to select the self learning mode. If A S L M is not set, the program flow goes to step S 7 to select the manual learning mode MLM. Then in step S 9 (Fig. 7(b)), a source sentence (SS) is inputted.
Thereafter, in steps S 10 to S 13, an automatic translation is executed in the machine by consulting a dictionary (C D), syntax analysis ~S A), conversion (CONV) of the source sentence to the translated sentence, and generation (GEN) of the translation sentence (in this case a Japanese sentence). Under the self learning mode or manual learning mode, the learned word stored in the memory 2a is used with first priority. The program flow goes to step S 14 to display the translated sentence (DISP). The operator judges whether or not the trans-lation is correct by viewing the display unit 3. An af-firmative conclusion is referred to as correct ? in step S 15. If the translation is not correct, it is judged in step S 23 whether the syntax analysis is correct or the translation of a word E~ se is correct ~referred to ~'~5~L5~3 as SAC or WDC)o If the syntax analysis is not correct, the program flow goes back to step S 11 by operation of a syntax analysis correction key, for the machine to perform the syntax analysis again. If only the word is not correct, the program flow goes to step S 24 to find a correct word (SWDC) by selecting any one of the Japanese words displayed in TABLE 1 seen in the display unit 3. A
correct translation can thus be displayed in the unit 3, the flow returning to step S ~4. When a correct trans-lation has been obtained, the program flow goes to stepS 16 to judge whether the self learning mode S L M is selected. If yes, it is judged in step S 17 whether or not the automatic self learning mode (A S L M) is set.
In the case of A S L M, the words learned tLW) in the translation process are stored in the self learning buf-fer 2a in step S 22. If A S L M is not set, the program flow goes to step S 18 to judge whether or not to learn the words used in the translation process. If learning is not desired, the program flow goes to step S 21 to judge whether there is a subsequent sentence to be trans-lated. If learning is desired, the progra~ flow goes to step S 19, wherein the words to be learned are designated by the operator. The words designated by the operator are then stored in the memory 2a in step S 20. If there are words to be cleared, such words can be cleared in steps S 25 and S 26. The program flow then goes to step S 27 to judge whether there is an instruction for a change of learning mode (referred to as CHM). If the learning mode is changed, the program flow goes to step S 2. If the learning mode is not changed, the program flow goes to step S 9 to prepare to receive the following translation.
A modification of a translation machine according to the present invention is shown in Fig. 8. In this modifi-cation there is provided a function of displaying a * mark on the word or words that are stored in the buffer by the :~25~
g learning operation, as shown in step S 28 in Fig. 8. This mode is abbreviated as D W D M in Fig. 8~ Also the term "learned word" is expressed as L W D. It is assumed that the following source English sentence is translated, I study air pollution in the field.
In this modification, if DWDM is set, the program flow goes to step S 28 wherein, at the time of generation of the translated sentence, if the translated sentence includes the learned word, the * mark is displayed on the display unit 3 ahead of the word in the manner shown below.
Watakushi wa taiki osen o (* genchi) de kenkyu suru.
In place of displaying the * mark the learned word can be emphasized by an underlini~g as follows.
Watakushi wa taiki osen o genchi de kenkyu suru.
A further modification is such that Watakushi wa taiki osen o 1. nohara de kenkyu suru.
2. hirogari 3. maizochi
Thus, every time the word "field" appears, the Japanese word "bunya" can be selected with first priority. If the operator wishes to delet~ the Japanese words alreadystored in the main memory, the unnecessary word can be deleted from the main memory ~ by use of a clear key in step S ~ In step S 5 it is judged whether or not an automatic self learning mode A S L M is set. I~ the A S L M is set, the program flow goes to step S 6 to select the self learning mode. If A S L M is not set, the program flow goes to step S 7 to select the manual learning mode MLM. Then in step S 9 (Fig. 7(b)), a source sentence (SS) is inputted.
Thereafter, in steps S 10 to S 13, an automatic translation is executed in the machine by consulting a dictionary (C D), syntax analysis ~S A), conversion (CONV) of the source sentence to the translated sentence, and generation (GEN) of the translation sentence (in this case a Japanese sentence). Under the self learning mode or manual learning mode, the learned word stored in the memory 2a is used with first priority. The program flow goes to step S 14 to display the translated sentence (DISP). The operator judges whether or not the trans-lation is correct by viewing the display unit 3. An af-firmative conclusion is referred to as correct ? in step S 15. If the translation is not correct, it is judged in step S 23 whether the syntax analysis is correct or the translation of a word E~ se is correct ~referred to ~'~5~L5~3 as SAC or WDC)o If the syntax analysis is not correct, the program flow goes back to step S 11 by operation of a syntax analysis correction key, for the machine to perform the syntax analysis again. If only the word is not correct, the program flow goes to step S 24 to find a correct word (SWDC) by selecting any one of the Japanese words displayed in TABLE 1 seen in the display unit 3. A
correct translation can thus be displayed in the unit 3, the flow returning to step S ~4. When a correct trans-lation has been obtained, the program flow goes to stepS 16 to judge whether the self learning mode S L M is selected. If yes, it is judged in step S 17 whether or not the automatic self learning mode (A S L M) is set.
In the case of A S L M, the words learned tLW) in the translation process are stored in the self learning buf-fer 2a in step S 22. If A S L M is not set, the program flow goes to step S 18 to judge whether or not to learn the words used in the translation process. If learning is not desired, the program flow goes to step S 21 to judge whether there is a subsequent sentence to be trans-lated. If learning is desired, the progra~ flow goes to step S 19, wherein the words to be learned are designated by the operator. The words designated by the operator are then stored in the memory 2a in step S 20. If there are words to be cleared, such words can be cleared in steps S 25 and S 26. The program flow then goes to step S 27 to judge whether there is an instruction for a change of learning mode (referred to as CHM). If the learning mode is changed, the program flow goes to step S 2. If the learning mode is not changed, the program flow goes to step S 9 to prepare to receive the following translation.
A modification of a translation machine according to the present invention is shown in Fig. 8. In this modifi-cation there is provided a function of displaying a * mark on the word or words that are stored in the buffer by the :~25~
g learning operation, as shown in step S 28 in Fig. 8. This mode is abbreviated as D W D M in Fig. 8~ Also the term "learned word" is expressed as L W D. It is assumed that the following source English sentence is translated, I study air pollution in the field.
In this modification, if DWDM is set, the program flow goes to step S 28 wherein, at the time of generation of the translated sentence, if the translated sentence includes the learned word, the * mark is displayed on the display unit 3 ahead of the word in the manner shown below.
Watakushi wa taiki osen o (* genchi) de kenkyu suru.
In place of displaying the * mark the learned word can be emphasized by an underlini~g as follows.
Watakushi wa taiki osen o genchi de kenkyu suru.
A further modification is such that Watakushi wa taiki osen o 1. nohara de kenkyu suru.
2. hirogari 3. maizochi
4. senjo
5. kyogijo
6. jimen
7. bunya
8. genchi
9. ba In the above modification, the learned word is the eighth 3apanese word "genchin.
If the DWDM is not set, the translated sentence is displayed without any mark in step S 29.
In judging whether or not the words are already learned, the CPU accesses the learning buffer 2a. If the translation word "genchi" of the source word "field" was already learned in the past, the word "genchi" is stored in the learning buffer 2a with the number 8 and "genchi"
as a pair. It can thus be judged that the word "genchi"
is the learned word.
5~7~3 Referring to Figs. 9 and lO(a) and lO(b) showing another embodiment of a translation machine according to the present invention, there is provided an external memory 7 using a RAM for storing the learned words in the form of a file called a learning file. Such learning file can comprise a plurality of files for storing the learned words for every field, for example, the words for mechani-cal engineering, chemical engineering and so on. In this embodiment, there are provided additional steps S 101 to S 103 between steps s 2 and s 3 shown in the flow chart of Fig. 7. In step S 103, it is judged whether or not the learning file 7 is called (referred to as "call L F 7" in Fig. lO(a)). This calling can be achieved by a specific key on the key board. When calling the learning file 7 is set, the program flow goes to step S 102 (shown as D L F) to designate the learning file 7. Then the program flow goes to step S 103 (shown as T L F) to transfer the con-tents of the learning file 7 into the learning buffer 2a.
Then the program flow goes to step S 7 and the translation operation already explained with reference to Fig. 7 is executed. If calling of the learning file 7 is not desig-nated, the program flow goes directly to step S 3.
After step S 21, there are provided additional steps S 106 and S 107. In step S 106 (shown as R L W) it is judged whether or not there is an instruction for regis-tering the learned words that are newly taught in the translation work into the learning file 7. If such re-gistration in the learning file 7 is indicated, the pro-gram flow goes to step S 107 (shown as R L W in L F 7) 3G to register the new learned words in the learning file 7 using the data stored in the learning buffer 2a. If such registration is not designated, the program flow goes directly to the E~D.
With the arrangement shown in this embodiment, since the words already learned are saved in the learning file 7(:~
7, these words can be used in every subsequent translation.
This means that, if a sentence to be translated is similar to a sentence already translated in the past, the work of the present translation can be performed using the learned words stored in the file 7 from the outset of the present work, thereby enabling the translation to proceed rapidly.
The learned words can be stored in separate files for every field.
Also, since the learned words are saved in the learning file 7, the same translated words can be used throughout a given work of translation.
If the DWDM is not set, the translated sentence is displayed without any mark in step S 29.
In judging whether or not the words are already learned, the CPU accesses the learning buffer 2a. If the translation word "genchi" of the source word "field" was already learned in the past, the word "genchi" is stored in the learning buffer 2a with the number 8 and "genchi"
as a pair. It can thus be judged that the word "genchi"
is the learned word.
5~7~3 Referring to Figs. 9 and lO(a) and lO(b) showing another embodiment of a translation machine according to the present invention, there is provided an external memory 7 using a RAM for storing the learned words in the form of a file called a learning file. Such learning file can comprise a plurality of files for storing the learned words for every field, for example, the words for mechani-cal engineering, chemical engineering and so on. In this embodiment, there are provided additional steps S 101 to S 103 between steps s 2 and s 3 shown in the flow chart of Fig. 7. In step S 103, it is judged whether or not the learning file 7 is called (referred to as "call L F 7" in Fig. lO(a)). This calling can be achieved by a specific key on the key board. When calling the learning file 7 is set, the program flow goes to step S 102 (shown as D L F) to designate the learning file 7. Then the program flow goes to step S 103 (shown as T L F) to transfer the con-tents of the learning file 7 into the learning buffer 2a.
Then the program flow goes to step S 7 and the translation operation already explained with reference to Fig. 7 is executed. If calling of the learning file 7 is not desig-nated, the program flow goes directly to step S 3.
After step S 21, there are provided additional steps S 106 and S 107. In step S 106 (shown as R L W) it is judged whether or not there is an instruction for regis-tering the learned words that are newly taught in the translation work into the learning file 7. If such re-gistration in the learning file 7 is indicated, the pro-gram flow goes to step S 107 (shown as R L W in L F 7) 3G to register the new learned words in the learning file 7 using the data stored in the learning buffer 2a. If such registration is not designated, the program flow goes directly to the E~D.
With the arrangement shown in this embodiment, since the words already learned are saved in the learning file 7(:~
7, these words can be used in every subsequent translation.
This means that, if a sentence to be translated is similar to a sentence already translated in the past, the work of the present translation can be performed using the learned words stored in the file 7 from the outset of the present work, thereby enabling the translation to proceed rapidly.
The learned words can be stored in separate files for every field.
Also, since the learned words are saved in the learning file 7, the same translated words can be used throughout a given work of translation.
Claims (7)
1. A translation system for translating sentences in a first language to sentences in a second language, comprising:
storage means for storing individual words in said first language together with a plurality of words in said second language, each word in said plurality of words being equivalent to the individual word stored with said plurality of words;
means for storing an input sentence in said first language to be translated;
means for determining the part of speech of each individual word in said stored input sentence;
means for selecting a first one of said plurality of words in said second language stored with an individual word as corresponding to said individual word based on the part of speech determination;
means for selecting another one of said plurality of words as corresponding to said individual word in response to a signal from a user indicating the previous selected word to be incorrect;
means for storing a selected word in said second language as a learned word corresponding to an individual word in said first language in response to a signal from a user indicating the correspondence to be correct; and means for selecting the learned word as said first one of said plurality of words in subsequent translations of input sentences containing said individual word.
storage means for storing individual words in said first language together with a plurality of words in said second language, each word in said plurality of words being equivalent to the individual word stored with said plurality of words;
means for storing an input sentence in said first language to be translated;
means for determining the part of speech of each individual word in said stored input sentence;
means for selecting a first one of said plurality of words in said second language stored with an individual word as corresponding to said individual word based on the part of speech determination;
means for selecting another one of said plurality of words as corresponding to said individual word in response to a signal from a user indicating the previous selected word to be incorrect;
means for storing a selected word in said second language as a learned word corresponding to an individual word in said first language in response to a signal from a user indicating the correspondence to be correct; and means for selecting the learned word as said first one of said plurality of words in subsequent translations of input sentences containing said individual word.
2. The translation system defined in claim 1, further comprising display means for displaying sentences in said first and second languages including means for displaying learned words with specified marks identifying said learned words.
3. The translation system according to claim 2, wherein said specific mark is an asterisk mark attached to the learned word.
4. The translation system according to claim 2, wherein said specific mark is a line to emphasize the learned word.
5. The translation system according to claim 2, wherein translated sentence is displayed on the display means and said mark is attached to a learned word contained in the displayed sentence.
6. The translation system according to claim 2, wherein said means for storing the learned word is a buffer which is cleared upon each start-up of said translation system.
7. The translation system according to claim 2, wherein said means for storing the learned word is an external memory means which can save learned words for subsequent start-ups of said translation system.
Applications Claiming Priority (6)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP103460/1985 | 1985-05-14 | ||
JP60103459A JPS61260367A (en) | 1985-05-14 | 1985-05-14 | Mechanical translating system |
JP60103460A JPS61260368A (en) | 1985-05-14 | 1985-05-14 | Translating system |
JP103459/1985 | 1985-05-14 | ||
JP60103458A JPS61260366A (en) | 1985-05-14 | 1985-05-14 | Mechanical translating system having learning function |
JP103458/1985 | 1985-05-14 |
Publications (1)
Publication Number | Publication Date |
---|---|
CA1251570A true CA1251570A (en) | 1989-03-21 |
Family
ID=27309995
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CA000509020A Expired CA1251570A (en) | 1985-05-14 | 1986-05-13 | Bilingual translation system with self intelligence |
Country Status (4)
Country | Link |
---|---|
US (1) | US4800522A (en) |
CA (1) | CA1251570A (en) |
DE (1) | DE3615972A1 (en) |
GB (1) | GB2177525B (en) |
Families Citing this family (50)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH02308370A (en) * | 1989-05-24 | 1990-12-21 | Toshiba Corp | Machine translation system |
US5418717A (en) * | 1990-08-27 | 1995-05-23 | Su; Keh-Yih | Multiple score language processing system |
GB9103080D0 (en) * | 1991-02-14 | 1991-04-03 | British And Foreign Bible The | Analysing textual documents |
US5541838A (en) * | 1992-10-26 | 1996-07-30 | Sharp Kabushiki Kaisha | Translation machine having capability of registering idioms |
US5303151A (en) * | 1993-02-26 | 1994-04-12 | Microsoft Corporation | Method and system for translating documents using translation handles |
JPH0916602A (en) * | 1995-06-27 | 1997-01-17 | Sony Corp | Translation system and its method |
ITUD980032A1 (en) * | 1998-03-03 | 1998-06-03 | Agostini Organizzazione Srl D | MACHINE TRANSLATION SYSTEM AND RESPECTIVE MACHINE TRANSLATION SYSTEM AND RESPECTIVE TRANSLATOR THAT INCLUDES THIS USER SYSTEM THAT INCLUDES THIS SYSTEM |
JP3696745B2 (en) | 1999-02-09 | 2005-09-21 | 株式会社日立製作所 | Document search method, document search system, and computer-readable recording medium storing document search program |
JP2001101187A (en) * | 1999-09-30 | 2001-04-13 | Sony Corp | Device and method for translation and recording medium |
US6848080B1 (en) * | 1999-11-05 | 2005-01-25 | Microsoft Corporation | Language input architecture for converting one text form to another text form with tolerance to spelling, typographical, and conversion errors |
US7403888B1 (en) * | 1999-11-05 | 2008-07-22 | Microsoft Corporation | Language input user interface |
DE10015858C2 (en) * | 2000-03-30 | 2002-03-28 | Gunthard Born | Process for computer-aided communication in natural languages related to semantic content |
DE10015859C2 (en) * | 2000-03-30 | 2002-04-04 | Gunthard Born | Process for computer-aided communication in natural languages based on grammatical content |
US7047493B1 (en) | 2000-03-31 | 2006-05-16 | Brill Eric D | Spell checker with arbitrary length string-to-string transformations to improve noisy channel spelling correction |
US7191115B2 (en) * | 2001-06-20 | 2007-03-13 | Microsoft Corporation | Statistical method and apparatus for learning translation relationships among words |
US8214196B2 (en) | 2001-07-03 | 2012-07-03 | University Of Southern California | Syntax-based statistical translation model |
JP2003030111A (en) * | 2001-07-19 | 2003-01-31 | Toshiba Corp | Mobile communication terminal device |
WO2004001623A2 (en) * | 2002-03-26 | 2003-12-31 | University Of Southern California | Constructing a translation lexicon from comparable, non-parallel corpora |
US20040044517A1 (en) * | 2002-08-30 | 2004-03-04 | Robert Palmquist | Translation system |
US7249012B2 (en) * | 2002-11-20 | 2007-07-24 | Microsoft Corporation | Statistical method and apparatus for learning translation relationships among phrases |
US7356457B2 (en) * | 2003-02-28 | 2008-04-08 | Microsoft Corporation | Machine translation using learned word associations without referring to a multi-lingual human authored dictionary of content words |
US7383542B2 (en) * | 2003-06-20 | 2008-06-03 | Microsoft Corporation | Adaptive machine translation service |
US8548794B2 (en) * | 2003-07-02 | 2013-10-01 | University Of Southern California | Statistical noun phrase translation |
US7711545B2 (en) * | 2003-07-02 | 2010-05-04 | Language Weaver, Inc. | Empirical methods for splitting compound words with application to machine translation |
CA2554890A1 (en) | 2003-12-17 | 2005-06-30 | Speechgear, Inc. | Translation tool |
US8296127B2 (en) * | 2004-03-23 | 2012-10-23 | University Of Southern California | Discovery of parallel text portions in comparable collections of corpora and training using comparable texts |
US8666725B2 (en) * | 2004-04-16 | 2014-03-04 | University Of Southern California | Selection and use of nonstatistical translation components in a statistical machine translation framework |
JP5452868B2 (en) * | 2004-10-12 | 2014-03-26 | ユニヴァーシティー オブ サザン カリフォルニア | Training for text-to-text applications that use string-to-tree conversion for training and decoding |
US8676563B2 (en) * | 2009-10-01 | 2014-03-18 | Language Weaver, Inc. | Providing human-generated and machine-generated trusted translations |
US8886517B2 (en) | 2005-06-17 | 2014-11-11 | Language Weaver, Inc. | Trust scoring for language translation systems |
US10319252B2 (en) * | 2005-11-09 | 2019-06-11 | Sdl Inc. | Language capability assessment and training apparatus and techniques |
US8943080B2 (en) | 2006-04-07 | 2015-01-27 | University Of Southern California | Systems and methods for identifying parallel documents and sentence fragments in multilingual document collections |
US8886518B1 (en) | 2006-08-07 | 2014-11-11 | Language Weaver, Inc. | System and method for capitalizing machine translated text |
US8433556B2 (en) | 2006-11-02 | 2013-04-30 | University Of Southern California | Semi-supervised training for statistical word alignment |
US9122674B1 (en) | 2006-12-15 | 2015-09-01 | Language Weaver, Inc. | Use of annotations in statistical machine translation |
US8468149B1 (en) | 2007-01-26 | 2013-06-18 | Language Weaver, Inc. | Multi-lingual online community |
US8831928B2 (en) * | 2007-04-04 | 2014-09-09 | Language Weaver, Inc. | Customizable machine translation service |
US8825466B1 (en) | 2007-06-08 | 2014-09-02 | Language Weaver, Inc. | Modification of annotated bilingual segment pairs in syntax-based machine translation |
US8990064B2 (en) | 2009-07-28 | 2015-03-24 | Language Weaver, Inc. | Translating documents based on content |
US8380486B2 (en) | 2009-10-01 | 2013-02-19 | Language Weaver, Inc. | Providing machine-generated translations and corresponding trust levels |
US10417646B2 (en) | 2010-03-09 | 2019-09-17 | Sdl Inc. | Predicting the cost associated with translating textual content |
US11003838B2 (en) | 2011-04-18 | 2021-05-11 | Sdl Inc. | Systems and methods for monitoring post translation editing |
US8694303B2 (en) | 2011-06-15 | 2014-04-08 | Language Weaver, Inc. | Systems and methods for tuning parameters in statistical machine translation |
US8886515B2 (en) | 2011-10-19 | 2014-11-11 | Language Weaver, Inc. | Systems and methods for enhancing machine translation post edit review processes |
US8942973B2 (en) | 2012-03-09 | 2015-01-27 | Language Weaver, Inc. | Content page URL translation |
US10261994B2 (en) | 2012-05-25 | 2019-04-16 | Sdl Inc. | Method and system for automatic management of reputation of translators |
US9152622B2 (en) | 2012-11-26 | 2015-10-06 | Language Weaver, Inc. | Personalized machine translation via online adaptation |
US9213694B2 (en) | 2013-10-10 | 2015-12-15 | Language Weaver, Inc. | Efficient online domain adaptation |
DE202016000135U1 (en) | 2016-01-08 | 2016-02-02 | Markus Robens | System for cooperatively improving language skills using terminology databases |
US10922496B2 (en) | 2018-11-07 | 2021-02-16 | International Business Machines Corporation | Modified graphical user interface-based language learning |
Family Cites Families (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS55147760A (en) * | 1979-05-08 | 1980-11-17 | Canon Inc | Electronic unit |
JPS5936795B2 (en) * | 1979-08-17 | 1984-09-05 | シャープ株式会社 | electronic dictionary |
JPS59869B2 (en) * | 1979-10-25 | 1984-01-09 | シャープ株式会社 | electronic translator |
JPS5858714B2 (en) * | 1979-11-12 | 1983-12-27 | シャープ株式会社 | translation device |
US4608665A (en) * | 1980-01-30 | 1986-08-26 | Kunio Yoshida | Sentence modifiable electronic dictionary and language interpreter |
US4623985A (en) * | 1980-04-15 | 1986-11-18 | Sharp Kabushiki Kaisha | Language translator with circuitry for detecting and holding words not stored in dictionary ROM |
JPS57152070A (en) * | 1981-03-13 | 1982-09-20 | Sharp Corp | Electronic interpreter |
JPS57172471A (en) * | 1981-04-17 | 1982-10-23 | Casio Comput Co Ltd | Searching system for electronic dictionary having extended memory |
DE3233194C2 (en) * | 1981-09-08 | 1986-04-10 | Sharp K.K., Osaka | Electronic pocket translator |
US4597056A (en) * | 1981-10-30 | 1986-06-24 | Sharp Kabushiki Kaisha | Language translator having circuitry for retrieving full words after single letter input |
JPS58175074A (en) * | 1982-04-07 | 1983-10-14 | Toshiba Corp | Analyzing system of sentence structure |
JPS59140583A (en) * | 1983-01-31 | 1984-08-11 | Sharp Corp | Electronic translating machine |
JPS6084667A (en) * | 1983-10-17 | 1985-05-14 | Mitsubishi Electric Corp | Sentence composing device |
JPS6126176A (en) * | 1984-07-17 | 1986-02-05 | Nec Corp | Dictionary for processing language |
DE3585937D1 (en) * | 1984-09-18 | 1992-06-04 | Sharp Kk | TRANSLATION SYSTEM. |
JPS61105671A (en) * | 1984-10-29 | 1986-05-23 | Hitachi Ltd | Natural language processing device |
JPH0664585B2 (en) * | 1984-12-25 | 1994-08-22 | 株式会社東芝 | Translation editing device |
JPS61217871A (en) * | 1985-03-25 | 1986-09-27 | Toshiba Corp | Translation processor |
-
1986
- 1986-05-13 US US06/862,323 patent/US4800522A/en not_active Expired - Fee Related
- 1986-05-13 DE DE19863615972 patent/DE3615972A1/en not_active Ceased
- 1986-05-13 CA CA000509020A patent/CA1251570A/en not_active Expired
- 1986-05-13 GB GB8611679A patent/GB2177525B/en not_active Expired
Also Published As
Publication number | Publication date |
---|---|
GB8611679D0 (en) | 1986-06-18 |
US4800522A (en) | 1989-01-24 |
DE3615972A1 (en) | 1986-11-20 |
GB2177525B (en) | 1989-08-16 |
GB2177525A (en) | 1987-01-21 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CA1251570A (en) | Bilingual translation system with self intelligence | |
US4814987A (en) | Translation system | |
KR900000094B1 (en) | Machinery translating devices | |
US4774666A (en) | Translating apparatus | |
US5005127A (en) | System including means to translate only selected portions of an input sentence and means to translate selected portions according to distinct rules | |
KR900008769B1 (en) | Machine interpretting system | |
KR900009120B1 (en) | Machine translation system | |
KR910008456B1 (en) | Machine translation processing apparatus | |
US4833611A (en) | Machine translation system | |
US4953088A (en) | Sentence translator with processing stage indicator | |
GB2241094A (en) | Translation machine | |
JP2815714B2 (en) | Translation equipment | |
EP0357344B1 (en) | Computer assisted language translating machine | |
US4860206A (en) | Translation system allowing user designation of postpositional words | |
US5257187A (en) | Translation machine system | |
US5075851A (en) | System for translating a source language word with a prefix into a target language word with multiple forms | |
GB2193018A (en) | Translation machine system | |
JPS61260366A (en) | Mechanical translating system having learning function | |
JPS6118074A (en) | Pre-editing system | |
JPS61260367A (en) | Mechanical translating system | |
JP3147947B2 (en) | Machine translation equipment | |
KR100424009B1 (en) | Translation Method and apparatus having function of displaying input errors | |
JP2806352B2 (en) | Dictionary maintenance equipment for machine translation | |
JP2723886B2 (en) | Machine translation apparatus and translation rule creation method | |
JPH0350668A (en) | Character processor |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
MKEX | Expiry |