JPH1097538A - Machine translation device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To surely display a finite sentence whose variable semanteme is not matched but which is approximated in terms of meaning against a sentence which a user inputs by precisely grasping a state where sentence constitution elements except for the variable semanteme are almost matched and the variable semanteme is not matched. SOLUTION: CPU 1 is divided into an analysis part 10, a retrieval part 11, a substitution part 12 and a control part 13 from the feature of an operation. The analysis part 10 analyzes input data recognized by an input recognition part. The retrieval part 11 retrieves a translation data base 7, a semanteme dictionary 8 and a variable semanteme dictionary 9, which are stored in ROM 6. The substitution part 12 substitutes the variable semanteme. The control part 13 executes control except for mentioned control such as the control of screen display. When the variable semanteme exists in the retrieved version and the origin corresponding to the variable semanteme is unique in the input of input cut, the variable semanteme is considered to be matched and matched degree is obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a machine translator for translating a conversation sentence in a first language into a second language by using a conversational example sentence registered in advance.

[0002]

2. Description of the Related Art Nowadays, as the number of machines that go abroad for sightseeing and the like increases, many examples of conversations that collect many conversational expressions have been published so as not to be troubled by conversations in foreign countries. Also,
There has been proposed a machine translation apparatus that digitizes these conversation examples and selects a conversation example by key operation or the like and displays a translation of the selected conversation example. For example, Japanese Patent Application No. 5
The invention of JP-A-6-147263 corresponds to this. Also,
In this machine translator, a variable part is provided in advance in a conversation example sentence, and even if some words of the input free sentence are replaced, the same example sentence is used. it can.

Further, a machine translator has been proposed which automatically selects an optimal example sentence for conversation by freely inputting a sentence to be translated and displays the translated sentence. This machine translator also has a variable unit, and by automatically replacing the input phrase, a sentence close to the input sentence is generated.
It is configured to output.

In this method, an example sentence for conversation and words that characterize each example sentence, grouped for each synonym,
I remember a lot. The former is called a fixed phrase, and the latter is called a semantic feature. A variable semantic feature is called a variable semantic feature.

In this method, semantic features are extracted from an input sentence, and a fixed sentence having a pattern of a combination thereof is searched. At this time, it is necessary to determine how much the input sentence and each fixed sentence match as a pattern of a combination of semantic features. In the related art, a method is employed in which the degree of coincidence between an input sentence and each fixed sentence is calculated by matching each word unit, and only a fixed sentence having a certain threshold or more is displayed.

[0006]

However, when the input sentence and a certain fixed sentence do not match only in variable semantic features and all other parts of the sentence match, if the degree of matching is calculated, the threshold value does not exceed the threshold. , The meaning may not be displayed even though it is very close. However, if a sentence that differs only in variable semantic features is displayed as a translation result, the user can replace the sentence by himself and convey it to the other party, and the communication opportunity will be further expanded.

[0007] As an example, it is assumed that there is an input of "I was separated from the tour operator." On the other hand, it is assumed that there is no fixed sentence having the same sentence, and the fixed sentence “I was separated from a friend” is the closest sentence. Here, it is assumed that the part corresponding to “friend” has a variable semantic feature, and the element “passenger” is not registered.

[0008] Even if the part of "train attendant" is different, if the translation of the expression "was broken" is known, there is a high possibility that the part can be used for communication. However, if the degree of coincidence between "I was separated from the tour conductor" and "I was separated from the friend" is calculated, it may be lower than the threshold value and may not be displayed in the conventional method. Therefore, the search result of "I was separated from a friend" is not obtained, and it is not possible to replace only the part of "friend" with "train attendant" and convey it.

In the present invention, the sentence constituent elements other than the variable semantic features almost match, and by correctly grasping the state where the variable semantic features do not match, the variable semantic features do not match. Display near fixed phrases.

[0010] In a travel scene, proper nouns such as place names and personal names are very frequently mentioned. Also,
Foreign place names and personal names are often displayed in Japanese in katakana. However, a part of a katakana character string may be misunderstood as a certain semantic feature, and a fixed sentence with a far different meaning may be displayed.

[0011] As an example, suppose there is an input "How much is San Francisco?" Here, it is assumed that there is no semantic feature “San Francisco”, and only a part of “Fran” is extracted as a variable semantic feature “money unit”. Then, the degree of matching with a fixed sentence such as "How much is it in franc" is high, and this sentence is displayed.

The present invention considers a character string which is different in character type from other parts of a sentence, such as katakana, and is not registered as a semantic feature as one unknown word. Is suppressed, and a sentence with a close meaning is surely displayed.

Similarly, a sentence including a proper noun such as a place name indicated by katakana is input, and although there is a fixed sentence in which all other parts except the proper noun are identical, Because the part does not match, the overall degree of matching is low, and the meaning is very close but not displayed.
In addition, a portion corresponding to a proper noun in a fixed sentence is often a variable semantic feature.

As an example, suppose there is an input "How much is up to San Francisco". On the other hand, there is a fixed phrase "How much is New York?"
"New York" is a "place name" variable semantic feature and does not include the element "San Francisco". These have obviously the same meaning, but when the degree of coincidence is calculated, they are lower than the threshold, and are not displayed by the conventional method.

According to the present invention, a character string that is different in character type from other parts of a sentence, such as katakana, and is not registered as a semantic feature is converted into an arbitrary variable semantic feature or a specific variable semantic feature. As a result, even in the case of an input of a proper noun or the like that is not registered as an element of the variable semantic feature, a sentence having a similar meaning is surely displayed.

In the case where a proper noun such as a person's name is input in the first language, it is rare that the proper noun is registered as an element of the variable semantic feature. Almost impossible.

As an example, assume that there is an input of "My name is Kimura". On the other hand, there is a fixed phrase "My name is Suzuki", but "Suzuki" is "person name"
If it is a variable semantic feature and does not include the element "Kimura", the user must replace the part corresponding to "Suzuki" with "Kimura" even if this sentence is displayed as a translation result. Inconvenient.

Therefore, in the present invention, a character string that is different in character type from other parts of a sentence, such as katakana, and is not registered as a semantic feature is used in a second language, such as Roman characters. By converting it into a dictionary and displaying it as a translation, proper nouns in the first language are also translated in a form suitable for the second language.

Furthermore, when a translation result is displayed, if a word that has not been matched and has not been input has been inserted, whether the displayed sentence is semantically similar to the input sentence Sometimes it is difficult to understand.

As an example, assume that "How much is a bus to San Francisco" is entered, and a fixed phrase "How much is a bus to New York" is retrieved. When this fixed phrase is displayed as a translation (search) result, the input sentence and the fixed phrase are “New York” and “San Francisco” at the beginning of the sentence.
And, although the text surface is very different and the sentence is very close, it is hard to understand at first glance.
There is a possibility of misunderstanding that sentences with similar meanings were not searched.

Therefore, in the present invention, when the variable semantic feature in the fixed sentence does not match the input sentence, a specific character or character string which is not an element of the variable semantic feature is inserted and displayed. It clearly indicates that the variable semantic features do not match, and makes it easy to understand whether other components of the sentence match.

Also, for example, when translating from Japanese to English
Although it is always expressed in the second language, such as titles such as Mr./Ms.,
If there is an input of a language expression that is not necessarily expressed in the first language, or if it appears as a second language, since there is no information in the first language on which word to use, an appropriate second language sentence Cannot be generated.

As an example, suppose there is an input of "Can you call Mr. Suzuki?" When translating this into English, it is necessary to specify whether it is "Mr. Suzuki" or "Ms. Suzuki". However, there is no information for determining the input in the input.

Therefore, in the present invention, an essential expression in the second language is also present as a variable semantic feature in the first language, and the user is allowed to select a correct element after the translation result is displayed, so that the user can always select the correct element. Generate expressions in line with second language culture.

[0025]

[Means for Solving the Problems]

<Means of Claim 1> A bilingual database in which example sentences of a conversation in a first language and sentences in a second language, which are translations thereof, are registered as a pair, and a semantic feature and a description extracted as the semantic feature. A semantic feature dictionary that defines and stores the same, a variable semantic feature dictionary that has the same semantic feature and registers replaceable words, and input means for inputting an input sentence for searching the bilingual database, Search means for searching the bilingual database, the semantic feature dictionary and the variable semantic feature dictionary based on the input sentence input from the input means, and the bilingual database and the semantic feature dictionary searched by the search means A replacement unit for performing replacement on the example sentence based on the variable semantic feature dictionary, an example sentence extracted by the search unit and the replacement unit, and an input sentence input from the input unit. An analysis unit that calculates a degree of coincidence with a power sentence, and an output unit that outputs a translation of an example sentence based on the notation analysis unit, wherein the analysis unit includes:
If a variable semantic feature is present in the retrieved translated sentence, a feature corresponding to the variable semantic feature does not exist in the input of the input means, and if there is only one semantic feature that does not match, the variable semantic feature is determined. The above problem is solved by obtaining the degree of coincidence as a coincidence.

In the analyzing means, the input sentence is a partial character string having a character type different from that of another part, and is registered as a semantic feature in the semantic feature dictionary and the variable semantic feature dictionary. The above problem is solved by recognizing a character string that does not exist as an unknown word and calculating the degree of matching.

<Means of Claim 3> In the analyzing means, the input sentence is a partial character string having a character type different from that of another part, and is registered as a semantic feature in the semantic feature dictionary and the variable semantic feature dictionary. The above problem is solved by recognizing a character string that does not exist as a specific variable semantic feature and calculating the degree of coincidence.

In the replacement means, the substitution means is a partial character string having a character type different from other parts in the input sentence, and is registered as a semantic feature in the semantic feature dictionary and the variable semantic feature dictionary. The above problem is solved by converting a character string that does not exist into a specific character type.

In the replacement means, if a character string corresponding to the variable semantic feature in the retrieved translated sentence does not exist in the input sentence, a specific character string is included in the variable semantic feature. The above-mentioned problem is solved by inserting and displaying.

In the replacement means, for a language expression which is always expressed in the second language but not necessarily expressed in the first language, the expression in the second language is set as a variable semantic feature, and The above problem is solved by giving and displaying the result.

[0031]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below in detail with reference to the drawings. In the following example, Japanese is used as the first language and English is used as the second language. Obviously, the language is not limited to these languages. FIG. 2 shows an appearance of a machine translation apparatus according to an embodiment of the present invention. Many pairs of Japanese and English sentence examples are stored, and these can be displayed on the liquid crystal display 2. This device uses handwritten character recognition with a dedicated pen 15 as an input interface. A display other than liquid crystal such as a CRT can be used, and other means such as voice or keyboard input can be used as input means.

FIG. 1 shows an outline of a circuit configuration of such a machine translation apparatus. The machine translator is a CPU (central processing unit) 1 for performing a central function of various controls.
Has been arranged. The CPU 1 sends the ROM 6, the working memory 5, the display driver 3, and the like through a bus 18 such as a data bus.
It is connected to each of the input recognition units 4.

The ROM 6 is a large-capacity read-only memory that stores programs for performing various controls of the machine translation device, a large number of phrases, and the like. The working memory 5 is a random access memory for temporarily storing data required for executing the program. The display driver 3 is a driver for displaying a bilingual sentence or the like on the liquid crystal display 2. The input recognition unit 4 recognizes input data from the handwriting input board 14 in FIG.

In the ROM 6, a bilingual database 7, a semantic feature dictionary 8, and a variable semantic feature dictionary 9 are stored. The bilingual database 7 is, for example, addresses 0 to 5000
0, the semantic feature dictionary 8 is stored as addresses 50001 to 70000, and the variable semantic feature dictionary 9 is stored as addresses 70001 to 100000.

The CPU 1 is divided into an analysis unit 10, a search unit 11, a replacement unit 12, and a control unit 13 according to the nature of the operation. The analysis unit 10 analyzes the input data recognized by the input recognition unit. The search unit 11 searches the bilingual database 7, the semantic feature dictionary 8, and the variable semantic feature dictionary 9 stored in the ROM 6. The replacement unit 12 replaces a variable semantic feature element.
The control unit 13 performs control other than the above, such as control of screen display.

An embodiment of the data structure of the bilingual database 7, the semantic feature dictionary 8, and the variable semantic feature dictionary 9 will be described below. These are common in the following description except for a part.

The bilingual database 7 defines a number of useful sentence examples that are desired to be translated and conveyed to local people when traveling overseas.

Example 1 How much is up to {New York}? How much is it to ｛New York｝? How much is ｛Place name｝? Example 2 How much is from ｛New York｝ to ｛Chicago｝? How much is it from ｛New York｝ $ 1 $ to ｛Chicago｝ $ 2 $? How much from ｛地名｝ to ｛地名｝? Here, the first line of the sentence example is the displayed Japanese sentence, and the second line is the translated sentence. Yes, the English sentence displayed together with the Japanese sentence as the translation result, and the third line is a set of semantic features associated with the bilingual sentence. This is called a semantic feature sequence. Also, the portion surrounded by ｛｝ is a variable semantic feature.

The semantic feature is a character string that becomes a keyword when searching for a sentence example, and is often assigned one to one sentence of each Japanese sentence. In “How much is it up to New York” in Example 1, “New York” and “until”
"How much" and "how much" are the semantic features of this sentence.
To be precise, these are the names of semantic features, and what character strings are defined as these semantic features are described in the semantic feature dictionary 8.

Further, "New York" in the sentence example 1 is a variable semantic feature. If various places such as "London" and "Oxford" can be entered in this part of the sentence example 1, the sentences that can be translated are doubled, which is very effective. Therefore, a Japanese character string that can enter this location and its translation are defined in the variable meaning feature dictionary 9, and the set is treated as one variable meaning feature called "place name".

In the bilingual database 7, the name of the variable semantic feature is described by enclosing it in the semantic feature sequence part, and one element thereof is written in parentheses in the Japanese sentence on the first line. Also, write the translated word in parentheses in the English sentence on the second line. Japanese text and English text are defined as parallel translations in the variable meaning feature dictionary 9. These also serve as representative values of the variable semantic feature in this sentence. That is, if there is no character string corresponding to the variable semantic feature in the input, this element is displayed.

For example, sentence example 1 has one variable semantic feature named "place name", and the representative value of the variable semantic feature in this sentence is "New York", and its translated word is "NewYork".

When there are a plurality of variable semantic features in one sentence,
The order of appearance of variable semantic features in the Japanese sentence matches the order of appearance of variable semantic features in the semantic feature sequence. Similarly, Sentence Example 2 has two variable semantic features named “place name”, and the representative value of the first “place name” variable semantic feature in this sentence is “New York”, and its translated word is “New York”.
The representative value of the second "place name" variable semantic feature is "Chicago"
And its translation is "Chicago".

Also, $ 1 $, $ 2 $ after the @ symbol in sentence 2
Indicates that the “place name” variable semantic attribute corresponds to the first and second variable semantic features in Japanese sentences, respectively.

That is, the first “place name” variable semantic feature including “New York” is ｛New York｝ $ 1 $,
The second "place name" variable semantic feature containing "Chicago" corresponds to "Chicago" $ 2 $.

When a semantic feature that is not variable is distinguished from a variable semantic feature, it is called an invariant semantic feature.

In the semantic feature dictionary 8, names of semantic features and Japanese character strings to be extracted as the semantic features are defined. For example, "from""to""howmuch""isit?"
Is defined in the semantic feature dictionary 8 as follows.

[Semantic feature dictionary 8] How much is from to? This means that the Japanese character string on the right is defined as the semantic feature of the name on the left. For example, the character string "how much" is defined as a semantic feature of the name "how much".

In the variable semantic feature dictionary 9, the name of the variable semantic feature, a Japanese character string as an element included therein,
The corresponding English translation is defined. For example, a set of a variable semantic feature element “place name” and its translation is defined in the variable semantic feature dictionary 9 as follows.

[Variable semantic feature dictionary 9] Place name New York Chicago Chicago London London Oxford Oxford This is a Japanese character string on the left (such as "New York")
Is defined as an element of the variable semantic feature "place name,
The translated word is the English string to the right of the string ("New Yor
k ”).

Hereinafter, the operation of the machine translation apparatus will be described. The following is a description of the processing common to the claims. The basic operation of this machine translation device is that a user inputs a translated sentence, detects a semantic feature from the sentence, detects a fixed sentence from the semantic feature, replaces and displays a variable semantic feature in the fixed sentence, That is.

Hereinafter, the translated sentence input by the user is referred to as an input sentence, and the Japanese sentence stored in the machine translator is referred to as a fixed sentence. When a pair of a fixed sentence and its translated English sentence is referred to, it is called a bilingual sentence.

The operation of the machine translation apparatus will be described with reference to the flowchart of FIG. The description will be given in a form showing details of each step of the flowchart of FIG. Take the process of translating the sentence "How much is it up to London" as an example (hereinafter this sentence is referred to as input sentence 1).

<Step A1> Controller 13 of CPU 1
Displays the “translated sentence input screen” of FIG. 4 on the liquid crystal display 14 through the display driver 3.

<Step A2> The user enters the dedicated pen 1 on the handwriting input board 16 on the translated sentence input screen shown in FIG.
With 5, input "hands up to London" by handwriting. FIG. 5 is a diagram in the middle of the input, and FIG. 6 is a state in which the input is completed.

<Step A3> The user presses the interpreter icon 17 in FIG. This input sentence is recognized by the input recognition unit 4 and stored in the work memory 5. Thereafter, the process proceeds to step A4.

<Step A4> Analysis section 10 of CPU 1
Is the input sentence 1 stored in the working memory 5 in step A3.
Then, a partial character string is created and stored in the working memory 5. Next, the process proceeds to step A5. The partial character strings of the input sentence 1 are as follows.

[0058] These are stored in the working memory 5.

<Step A5> Search unit 11 of CPU1
Searches the semantic feature dictionary 8 and the variable semantic feature dictionary 9 stored in the ROM 6 using the partial character string of the input sentence 1 stored in the working memory 5 in step A4 as a key.
If there is a character string defined as a semantic feature in the semantic feature dictionary 8 among the partial character strings, the search unit 11 of the CPU 1 stores the character string and the name of the semantic feature in the working memory 5 as a pair, and proceeds to step A6. Move on to If not, the process proceeds to step A16.

At this time, the partial character strings are searched in order from the front, and the order is stored, so that the appearance order of the input sentences of each semantic feature is also stored.

Hereinafter, a process of searching the semantic feature dictionary 8 and the variable meaning attribute dictionary 9 using the partial character string "how much" of the input sentence 1 as a key will be described.

The search unit 11 of the CPU 1 reads the semantic feature dictionary 8 stored in the ROM 6 line by line from address 50001, and stores it in the work memory 5. Each row of the semantic feature dictionary 8 describes the name of each semantic feature on the left side, and a character string defined as the semantic feature on the right side. Assume that the semantic feature dictionary 8 has the following lines.

[Semantic feature dictionary 8] Now, since the character string “how much” is defined in the semantic feature dictionary 8, the search unit 11 of the CPU 1 searches the semantic feature name “how much” on the left side. Then, the partial character string “how much” of the input sentence is stored in the working memory 5 as a pair.

Since the character string "how much" is not defined in other places in the semantic feature dictionary 8, the search of the semantic feature dictionary 8 is terminated.

Next, the search unit 11 of the CPU 1 searches the variable semantic feature dictionary 9 using "how much" as a key. Since the character string "how much" does not appear in the variable semantic feature dictionary 9, the search using "how much" as a key ends, and the next partial character string is set as a key.

The process of searching the semantic feature dictionary 8 and the variable semantic feature dictionary 9 using "London" as a key will be described below. Now, assume that the character string “London” is stored in the working memory 5.

The search unit 11 of the CPU 1 checks whether a character string “London” is defined in the semantic feature dictionary 8 stored in the ROM 6. Now, since the character string “London” does not appear in the semantic feature dictionary 8,
Next, the variable semantic feature dictionary 9 is searched. It is assumed that the variable semantic feature dictionary 9 has the following description.

[Variable Semantic Feature Dictionary 9] The place name New York Chicago Chicago London London Oxford Oxford Since the character string “London” is defined in the variable meaning feature dictionary 9, the retrieval unit 11 of the CPU 1 includes the variable The name of the semantic feature "place name" and the translation of the element "Londo
n ”is stored in the working memory 5.

Since the character string "London" is not defined elsewhere in the variable meaning feature dictionary 9, the search of the variable meaning feature dictionary 9 is terminated.

The above operation is performed for all partial character strings of the input sentence 1. It is assumed that the following variable / invariant semantic features are detected from the input sentence 1.

[Semantic feature corresponding to partial character string] "London" = "place name": variable semantic feature "to" = "to": invariant semantic feature "How much" = "how much": invariant semantic feature "is" = “Is”: invariant semantic feature Here, the right side of = is the name of the semantic feature, and the character string on the left side is detected as a variable / invariant semantic feature of the name on the right side. Also, the translation “London” of the variable semantic feature element “London” is stored.

[Variable semantic element] "London" = "London" Here, the right side of = is a translation of the corresponding element of the variable semantic feature, and the words on the right side are detected as the translations of the words on the left side, respectively. means. Further, it is assumed that each semantic feature detected from the input sentence is detected in the order from top to bottom in the above correspondence table. This order is stored in the working memory 5. This concludes the description of step A5 according to the example. In this example, since the semantic feature was searched, the next step A6
Move on to

<Step A6> Search unit 11 of CPU 1
Searches the bilingual database 7 stored in the ROM 6 using the name of the variable / invariant semantic feature stored in the working memory 5 in step A5 as a key. If any of the variable / invariant semantic feature names appear in the bilingual database 7, the fixed form sentence in which the semantic feature appears is stored in the working memory 5 together with the English sentence of the translated sentence and the semantic feature sequence. Move to A7. If not, the process proceeds to step A16.

The details of step A6 will be described using the input sentence 1 as an example. First, the CPU 1 searches the bilingual database 7 using one of the semantic feature sets stored in the working memory 5 at step A5 as a key. First, the bilingual database 7 is searched using the semantic feature “place name” as a key. It is assumed that the bilingual database 7 has the following description.

[Bilingual translation database 7] In sentence example 1, the “sequence name” semantic feature is included in the semantic feature sequence. Therefore, the search unit 11 of the CPU 1 stores the sentence example 1 in the working memory 5. In the bilingual database 7,
Other places do not include "place name" semantic features. Thus, the search using “place name” as a key is terminated.

The search unit 11 of the CPU 1 performs the above-mentioned work for all the semantic attributes stored in the work memory 5. Now, it is assumed that the following sentence examples 1 and 3 are stored in the working memory 5 after the search for all semantic features is completed.

[0077] Sentence example 3 was searched by the semantic feature “how much”. This is the end of the description of step A6 according to the example. In this example, since a fixed phrase has been searched, the process proceeds to step A7.

<Step A7> Analysis section 10 of CPU 1
Selects one sentence stored at the head among the fixed sentences stored in the working memory 5 in step A6. The selected fixed phrase is flagged. Here, sentence example 1 is selected and a flag is set.

<Step A8> Analysis section 10 of CPU 1
Calculates the degree of coincidence between the fixed sentence selected in step A7 or A12 and the input sentence. Here, the calculation is performed using the number of the semantic features of the fixed sentence that match the semantic features in the input sentence. Then, the pair of the fixed phrase and the degree of coincidence is stored in the working memory 5, and if the degree of coincidence is equal to or more than a certain threshold, the process proceeds to step A9.
Move to 3. Here, the threshold is set to 200 points. Now, sentence example 1 is selected.

[0080] Semantic features of sentence example 1 "place name""until""howmuch""isit?"
Are all in the input sentence. The number of matching sentences is multiplied by 100, and the matching degree between the fixed sentence and the input sentence is set to 400 points. Therefore, the sentence example 1 and the matching score of 400 are stored in the working memory 5. Since the matching score of 400 in the example 1 exceeds the threshold value of 200, the process proceeds to step A9.

<Step A9> Analysis section 10 of CPU 1
Determines whether the fixed phrase stored in the working memory 5 in step A8 has a variable semantic feature. If there is, the process proceeds to step A10, and if not, the process proceeds to step A11.

Since the sentence example 1 has the variable semantic feature "place name", the process proceeds to step A10.

<Step A10> Replacement unit 12 of CPU 1
Replaces the variable meaning attribute element in the Japanese sentence of the fixed sentence stored in the working memory 5 with the variable meaning feature element in the input sentence 1. Further, the variable meaning attribute element in the English sentence is replaced with a translation of the element stored in the working memory 5. The bilingual sentence is stored in the working memory 5 in the replaced form. Thereafter, the process proceeds to step A11.

Now, the element “New York” is included in the “place name” variable semantic feature of the sentence example 1. On the other hand, the “place name” variable semantic feature element detected from the input sentence 1 is “London”
It is. At the same time, the translation “London” of “London” is also stored in the working memory 5.

Therefore, the replacement unit 12 of the CPU 1 replaces “New York” in the first example with “London” and replaces “New Yo
Replace "rk" with "London". It looks like this:

[0086] The data is stored in the working memory 5 in this manner, and the process proceeds to Step A11.

<Step A11> Analysis section 10 of CPU1
Checks whether there is a fixed sentence with no flag in the working memory 5. Step A12 if present
If not, the process proceeds to step A14. Since no flag is set in the sentence example 3, the process proceeds to step A12.

<Step A12> Analysis section 10 of CPU 1
Selects one sentence in the working memory 5 on which no flag is set and sets a flag. Then, step A8
Move on to Here, sentence example 3 is selected, a flag is set, and the routine proceeds to step A8.

<Step A8> Analysis section 10 of CPU 1
Calculates the degree of coincidence of Sentence Example 3.

Example 3: What is the exchange rate today? What's the exchange rate today? How much is the exchange rate today?
"Is it?" The remaining “today”, “exchange”, and “rate” do not exist in the input sentence. Further, the two semantic features “place name” and “until” in the input sentence do not exist in sentence example 3. Therefore, minus 50, which is obtained by multiplying the number of matching 2 by 100 and subtracting the number of unmatching 5 by 50, is defined as the matching degree of the fixed sentence with the input sentence.
Therefore, the sentence example 3 and the degree of coincidence of −50 are stored in the working memory 5. Since the matching score of −50 in Sentence Example 3 is lower than the threshold value of 200, the process proceeds to Step A13.

<Step A13> Analysis section 10 of CPU1
Deletes, from the working memory 5, the fixed phrase having the degree of coincidence less than the threshold value stored in the working memory 5 in step A8. Then, the process proceeds to step A11.

Sentence example 3 in which the degree of coincidence is -50 points
The data is erased from the working memory 5, and the process proceeds to Step A11.

<Step A11> Analysis section 10 of CPU 1
Detects that there is no fixed phrase with no flag set in the working memory 5, and proceeds to step A14.

<Step A14> Controller 13 of CPU 1
Switches the liquid crystal display 14 from the free text interpretation input screen of FIG. 6 to the bilingual text display screen of FIG. 7 through the display driver 3. Thereafter, the process proceeds to step A15.

<Step A15> Controller 13 of CPU 1
Displays the Japanese sentence and the English sentence of the sentence example stored in the work memory 5 on the liquid crystal display 14 through the display driver 3 as shown in FIG. At this time, the variable semantic features are underlined and displayed. In addition, when displaying, fixed phrases with a high degree of matching are displayed in order. Now, working memory 5
Is only sentence example 1. The Japanese and English sentences are displayed as shown in FIG.

The description of the operation of the machine translator using the input sentence 1 as an example is now completed. Next, FIG. 3 which was not traced above
Steps in the flowchart will be described.

If no semantic feature or fixed phrase is found in step A5 or A6, it is determined that translation is not possible and that fact is displayed. The processing is described in step A16 below.
It is.

<Step A16> Control unit 13 of CPU 1
Switches the liquid crystal display 14 from the free text interpretation input screen of FIG. 6 to the non-translatable screen of FIG. 9 through the display driver 3. The above is the common operation of the machine translation device of the present invention.

In this embodiment, the retrieved sentence is displayed as a translation, but only a Japanese sentence or an English sentence may be displayed. Also,
Drawings are examples of the design of the screen to be displayed, and various designs are conceivable depending on the application.

In this embodiment, the degree of coincidence between an input sentence and a fixed sentence is obtained by multiplying the number of matching semantic features by 100 to the number of non-matching semantic features by 50. Although a subtracted number is used, various methods other than this, such as dividing by the number of semantic features in the input sentence to obtain a relative evaluation of 100 points or more, can be considered.

Embodiments will be described below for each claim. In each embodiment, as in the above-described basic embodiment, a process of displaying a bilingual sentence for a certain input sentence will be described as an example.

Hereinafter, an embodiment of the machine translation apparatus according to claim 1 will be described. The process of translating an input sentence by the machine translation device according to claim 1 is the same as that of the above-described basic embodiment except for step A8. Therefore, only the difference in step A8 will be described in detail here. Since the operation flow is the same as that of the basic embodiment, the operation will be described with reference to the flowchart of FIG.

Now, in steps A1 and A2, the user inputs "I was separated from the tour operator", and in steps A3 and A2
Assume that the following sentence example is searched in A6.

[0104] Here, it is assumed that the “person” variable semantic feature does not have an element of “guest”. At this time, step A of the basic embodiment
When the degree of coincidence with the input sentence of sentence example 4 is calculated based on No. 8,
Of the semantic features of Sentence Example 4, "person", "separate" and "taita", "separate" and "taita" are in the input sentence. The remaining "person" variable semantic features do not exist in the input sentence. In addition, the sentence example 4 does not include the semantic feature of “passenger” in the input sentence.

Therefore, the degree of coincidence with the input sentence of Example 4 is 100 points obtained by subtracting the result of multiplying the number of unmatched numbers 2 by 100 and the number of unmatched numbers 2 multiplied by 50. . However, the score of 100 in Example 4 is the threshold 2
Since it is lower than 00, it is erased in the next step A13.

In the present invention, a fixed sentence that is similar to the input sentence semantically similar to the input sentence, except that only the portion corresponding to the variable semantic feature is calculated, is simply calculated as a mismatch as described above. Instead, a special process is performed to ensure display.

According to the present invention, in step A8 of the basic embodiment, if "the variable semantic feature exists in the fixed sentence, the corresponding element does not exist in the input sentence, and the semantic feature of the fixed sentence is included in the input sentence. If there is only one semantic feature that does not match with, the variable semantic feature that does not match in the fixed sentence and the semantic feature that does not match in the input sentence are regarded as being matched are calculated. I do.

Now, in sentence example 4, there is a variable semantic feature "person" that does not match the input sentence. Also, in the input sentence, there is only one “passenger” whose semantic feature does not match the semantic feature of the fixed phrase. Therefore, the calculation is performed on the assumption that the “person” and the “passenger” coincide.

Then, sentence example 4 and the input sentence have all the same semantic features. By multiplying the number 3 of matches by 100, the degree of matching with the input sentence of Sentence Example 4 is 300 points.
Sentence example 4 and the matching score of 300 are stored in the working memory 5.
Since the matching score of 300 in the example 4 exceeds the threshold value of 200, the process proceeds to step A9.

Subsequent steps are the same as in the basic embodiment. In steps A9 to A15, the sentence example 4 is displayed on the liquid crystal display 4 as shown in FIG. 15, and the series of translation operations ends.

In this embodiment, the "passenger" is registered as a semantic feature. Even when the "passenger" is not registered, the "person" variable semantic feature is set as an unknown word. Can also be matched.

Next, an embodiment of the machine translation apparatus according to claim 2 will be described. In the present invention, a character string which is not registered as a semantic feature in a subsequence having a character type different from that of another part of the sentence is regarded as one unknown word, but in the present embodiment, a katakana character string is taken as an example. explain.

The process of translating an input sentence by the machine translation apparatus according to the second aspect is the same as that of the basic embodiment except for steps A4 and A5. Therefore, steps C1 to C5 which replace steps A4 and A5 will be described with reference to the flowchart of FIG. Other steps are the same as those in FIG.

Now, it is assumed that the user inputs "How much is up to San Francisco" and presses the interpreter icon 17 in steps A1 to A3. Next, the process proceeds to step C1. Here, if the processing is performed according to A4 and A5 of the basic embodiment, a part of the character string "San Francisco" is extracted as a semantic feature, and the meaning of "Fran how much is it?" Search for fixed phrases. The following steps C1 to C5 are processes for preventing such a situation.

<Step C1> Analysis section 10 of CPU1
Is the input sentence 1 stored in the working memory 5 in step A3.
About the character type of each character. If there is katakana in the input sentence, the process proceeds to step C2, otherwise to step A4. Now, the input sentence "How much is San Francisco?"
Contains katakana, the process proceeds to step C2.

<Step C2> Analysis section 10 of CPU 1
Is the input sentence 1 stored in the working memory 5 in step A3.
Then, a partial character string is created and stored in the working memory 5. Next, the process proceeds to step C3. At this time, a partial character string that starts in the middle of the katakana character string is not created. In addition, a flag is set for a partial character string including only katakana.

Now, "How much is it up to San Francisco" includes the katakana line "San Francisco", so "It is up to Francisco" or "How much is it up to Francisco"
Do not create substrings such as A flag is set for the partial character string "San Francisco".

<Step C3> Search unit 11 of CPU1
Searches the semantic feature dictionary 8 and the variable semantic feature dictionary 9 stored in the ROM 6 using the partial character string of the input sentence 1 stored in the working memory 5 in step C2 as a key.
If there is a character string defined as a semantic feature in the semantic feature dictionary 8 among the partial character strings, the search unit 11 of the CPU 1 stores the character string and the name of the semantic feature in the working memory 5 as a pair, and proceeds to step A6. Move on to If not, the process proceeds to step A16. At this time, the translated word of the variable semantic feature element is also stored in the working memory 5.

Now, it is assumed that semantic features such as "until", "how much", and "how much" are retrieved, and no semantic features or variable semantic features are retrieved from the character string "San Francisco".

<Step C4> Analysis section 10 of CPU 1
Checks whether a semantic feature or a variable semantic feature has been retrieved in step C3 from the katakana character string flagged in step C2. If it is found, the flag standing in the katakana character string is deleted, and step A6 is executed.
Move on to If not found, the process proceeds to step C5.

Now, in step C2, the character string "San Francisco" is flagged, and no semantic feature or variable semantic feature is retrieved from "San Francisco". Therefore, the process proceeds to Step C5 without deleting the flag.

<Step C5> Analysis section 10 of CPU 1
Replaces the katakana character string with the flag with "unknown word"
These are stored in the working memory 5 in pairs as semantic features of the names. Next, the process proceeds to step A6.

Now, a flag is set for "San Francisco". Therefore, “San Francisco” is a semantic feature of “unknown word”, and “San Francisco” and the name of the semantic feature “unknown word” are stored in the working memory 5. Thereafter, the process proceeds to step A6.

Subsequent steps are the same as in the basic embodiment. Assume that the following sentence example is searched in step A6.

Example sentence 1: How much is up to {New York}? How much is it to ｛New York｝?｝ Place name ｛How much is it? Here, it is assumed that there is no “San Francisco” element in the “place name” variable semantic feature.

Next, the degree of coincidence is calculated in step A8. The input sentence and the fixed phrase have the same three semantic features of “until”, “how much”, and “do”, and the two semantic features of “place name” and “unknown word” do not match. Therefore, the degree of coincidence is 20
It becomes 0 point and is not erased. Therefore, in step A15,
Sentence example 1 is displayed on the liquid crystal display 14 as shown in FIG.

As a result, the possibility of misunderstanding the part "Fran" of the character string "San Francisco" as the French monetary unit "Fran" and completely retrieving a sentence relating to money is completely eliminated.

Next, an embodiment of the machine translation apparatus according to claim 3 will be described. The claim is to consider a character string that is not registered as a semantic feature in a substring having a character type different from that of other parts of the sentence as an arbitrary variable semantic feature. To explain.

The process of translating an input sentence by the machine translation apparatus according to the third aspect excludes steps C3 and A8.
This is the same as the embodiment. Therefore, here, step A
8 and the step D5 replacing the step C3 will be described with reference to the flowchart of FIG. Other steps are the same as those in FIG.

Now, it is assumed that the user inputs "How much is up to San Francisco" and presses the interpreter icon 17 in steps A1 to A3. Next, steps C1 to C4
In the search, the input sentence is searched for semantic features such as "to", "how much", "is", and a flag is set in the katakana character string "San Francisco". The following step D5 replaces step C5 of claim 2.

<Step D5> Analysis section 10 of CPU 1
Stores the katakana character string on which the flag is set as a variable semantic feature with a name of “arbitrary” in the working memory 5 as a pair. Further, the null character string is stored in the working memory 5 as a translation, and the process proceeds to step A6.

Now, a flag is set on the character string "San Francisco". Therefore, “San Francisco” is a variable semantic feature “arbitrary”, and the name “arbitrary” of “San Francisco” and the variable semantic semantic feature is stored in the working memory 5. Further, the empty character string is stored as a translation, and the process proceeds to step A6. Steps A6 and A7 are the same as the above processing.

In step A6, a fixed form sentence is searched from the semantic features extracted from the input sentence. Suppose that the next sentence was searched from the input sentence "How much is up to San Francisco".

[0134] In step A7, this sentence is selected, and in step A8
Is calculated. Now, the input sentence and sentence example 1 have the same semantic features of “until”, “how much”, and “do”.

Here, in the present invention, in step A8 of the embodiment (basic embodiment) of claim 2, "if an input sentence has a variable semantic feature with a name of" arbitrary ", it will Is considered to be consistent with the variable semantic feature of "."

Now, the input sentence has an “arbitrary” variable semantic feature, and the sentence example 1 has a variable semantic feature “place name”. Thus, all four match, and the number of matches is 100
, And the degree of coincidence of the fixed sentence with the input sentence is set to 400 points. Therefore, 400 points of coincidence with sentence example 1 are stored in working memory 5.
To be stored.

Since the matching score of 400 in the example 1 exceeds the threshold value of 200, the process proceeds to step A9. Here, the “arbitrary” variable semantic feature is calculated as matching with the arbitrary variable semantic feature of the fixed phrase, but the processing is performed only in this step. Therefore, in step A10, the element replacement of the variable semantic feature that matches the “arbitrary” variable semantic feature is not performed.

Subsequent steps are the same as in the second embodiment (basic embodiment). In steps A9 to A15,
Sentence example 1 is displayed on the liquid crystal display 4 as shown in FIG. 16, and a series of translation operations ends.

Further, a character string which is not registered as a semantic feature in a substring having a character type different from that of another part of the sentence may be regarded as a specific variable semantic feature. For example, a katakana proper noun in a scene of a trip is considered to represent a place name or a person name in most cases, and is regarded as a variable semantic feature such as “place name” or “person name”.

An example will be described below. Except for step C5, it is the same as the second embodiment. Therefore, here, step E5 replacing step C5 will be described with reference to the flowchart in FIG. Other steps are the same as those in FIG.

Now, it is assumed that the user inputs "How much is up to San Francisco" and presses the interpreter icon 17 in steps A1 to A3. Next, steps C1 to C4
In the search, the input sentence is searched for semantic features such as "to", "how much", "is", and a flag is set in the katakana character string "San Francisco". The next step E5 replaces step C5 of claim 2.

<Step E5> Analysis section 10 of CPU 1
Considers the katakana character string with the flag set as a variable semantic feature of the names "place name" and "person name", and stores these names and katakana character strings in the working memory 5 in pairs. Further, the null character string is stored in the working memory 5 as a translation, and the process proceeds to step A6.

Now, a flag is set on the character string "San Francisco". Therefore, “San Francisco” is a “place name” and “person name” variable semantic feature, and “San Francisco” and a variable semantic feature name “place name”
The “person name” is stored in the work memory 5. Further, the empty character string is stored as a translation, and the process proceeds to step A6. Subsequent steps are the same as in the basic embodiment. It is assumed that the following sentence example is searched in step A6.

[0144] Since all the semantic features of the input sentence and sentence example 1 match, the degree of matching is 400 points. Accordingly, in step A15, the sentence example 1 is displayed on the liquid crystal display 14 as shown in FIG. 16, and the series of translation operations ends.

Next, an embodiment of the machine translation apparatus according to claim 4 will be described. The claim is to convert a character string that is not registered as a semantic feature in a subsequence having a character type different from that of the other parts of the sentence to a character type used in a second foreign language. A case where a katakana character string is converted into an alphabet (Roman character) will be described as an example. Here, based on the embodiment of the machine translation apparatus of claim 3, only differences from the embodiment will be described.

In one embodiment of the present invention, information for converting katakana into Roman characters is stored in the variable semantic feature dictionary 9. This is a correspondence between all katakana characters and their romaji characters as follows.

[Variable Meaning Feature Dictionary 9] a a i i: The operation of claim 4 will be described below. The process of translating an input sentence by the machine translation device according to the fourth aspect is the same as that of the third embodiment except for one step. Therefore,
Only the added steps will be described with reference to the flowchart in FIG. Other steps are the same as those in FIG.

Now, in steps A1 to A3 and C1 to C2, the user inputs "My name is Kimura",
It is assumed that a flag is set in the katakana column "Kimura", which is stored as "place name" and "person name" variable semantic features. Further, it is assumed that the next fixed sentence is searched from the input sentence.

[0149] These fixed phrases and input sentences are "I", "name", "person name", "is"
All four match. Therefore, in step A9, 400 points obtained by multiplying the number of coincidences 4 by 100 are the coincidences. Since the coincidence is equal to or more than the threshold 200, the points are not erased.

In the present claim, the following step is added immediately before step A10 of the embodiment of the basic claim 3 (basic embodiment).

<Step F1> Analysis section 10 of CPU1
Converts all katakana in the character string into Roman characters with reference to the variable semantic feature dictionary 9 for the katakana string on which the flag stored in the working memory 5 is set. At this time, the first character is capitalized. The converted character string is stored in the working memory 5 as a translation of this element, and the process proceeds to step A10.

Now, a flag is set in the katakana column "Kimura". Refer to the variable semantic feature 9 and refer to “kimu
"r" is converted to the Roman time, and the first character is capitalized and stored in the working memory 5 as "Kimura".

Subsequent steps are the same as the third embodiment (basic embodiment). In step A10, “Suzuki” in sentence 8 is replaced with “Kimura”, and “Suzuki” is replaced with “Kimu
ra ”. In practice, it looks like this:

[0154] In steps A11 to A15, the sentence example 8 is displayed on the liquid crystal display 14 as shown in FIG. 11, and the series of translation operations ends.

Hereinafter, an embodiment of the machine translation apparatus according to claim 5 will be described. The present invention is to insert and display a certain character or character string in the variable semantic feature when there is no input of a character string corresponding to the variable semantic feature of the fixed phrase. A description will be given of an example of a case where “〜” is inserted and displayed.

The process of translating an input sentence by the machine translation device according to claim 5 is the same as that of the basic embodiment except for two steps G1 and G2. Therefore, only the added steps will be described in detail with reference to the flowchart of FIG. Other steps are the same as those in FIG. Now, it is assumed that the user inputs "Can you call Mr. Kimura?", And the sentence example 5 in the bilingual database 7 is searched.

[Bilingual translation database 7] Also, it is assumed that the “person name” variable semantic feature does not have the element “Kimura”.

In this fixed phrase and the input phrase, "san", "call" and "would you get" match, and "person name"
“Kimura” does not match. Therefore, the number of unmatched numbers 3 multiplied by 100 is subtracted by the number of unmatched numbers 2 multiplied by 50, and the degree of matching is 200 points. This is not erased because it is equal to or greater than the threshold of 200 points.

Next, in step A10, variable semantic feature replacement is performed. However, since a character string corresponding to the "person name" variable semantic feature in sentence example 5 does not exist in the input sentence, no replacement is performed. In this claim, step A11 of the basic embodiment
Add the following two steps just before.

<Step G1> Analysis section 12 of CPU1
Is a step A in the fixed phrase stored in the working memory 5.
It is checked whether or not there is a variable semantic feature that was not replaced by 10. If it exists, the process proceeds to step G2; otherwise, the process proceeds to step A11.

Now, the “person name” variable semantic feature in sentence 5 is
Not replaced in step A10. Therefore Step G
Move to 2.

<Step G2> Replacement unit 12 of CPU1
Inserts the character string “〜” into the portion of the fixed sentence stored in the working memory 5 corresponding to the variable semantic feature not replaced in step A10 in the Japanese sentence / English sentence. In that form, it is stored in the working memory 5, and the process proceeds to step A11.

In the sentence example 5, “Suzuki” corresponding to the “person name” variable semantic feature in the Japanese sentence, and “Mr. Su” in the English sentence.
Replace "zuki" with "~". As a result:

[0164] The data is stored in the working memory 5 in this manner, and the process proceeds to Step A11.

The subsequent steps are the same as in the basic embodiment. In steps A11 to A15, the sentence example 5 is displayed on the liquid crystal display 14 as shown in FIG. 12, and the series of translation operations ends.

When a character string corresponding to a variable semantic feature of a fixed phrase is not input, a name (upper concept) describing an included element may be inserted and displayed. For example,
A description will be given of an example in which the name of the variable semantic feature is inserted and displayed.

Now, the user inputs "Can you call Mr. Kimura?" And the sentence example 5 in the bilingual database 7 is entered.
Is searched.

[Bilingual Translation Database 7] Also, it is assumed that the “person name” variable semantic feature does not have the element “Kimura”.

In the fixed phrase and the input phrase, "san", "call" and "would you get" match, and the "person name"
“Kimura” does not match. Therefore, the number of unmatched numbers 3 multiplied by 100 is subtracted by the number of unmatched numbers 2 multiplied by 50, and the degree of matching is 200 points. This is not erased because it is equal to or greater than the threshold of 200 points.

Next, in step A10, variable semantic feature substitution is performed. However, since a character string corresponding to the "person name" variable semantic feature in sentence example 5 does not exist in the input sentence, no substitution is performed. The following two steps are added immediately before step A11 in the basic embodiment (FIG. 21).

<Step H1> Analysis section 12 of CPU1
Is a step A in the fixed phrase stored in the working memory 5.
It is checked whether or not there is a variable semantic feature that was not replaced by 10. If it exists, the process proceeds to step H2; otherwise, the process proceeds to step A11.

Now, in Example 5, the “person name” variable semantic feature is
Not replaced in step A10. Therefore, step H
Move to 2.

<Step H2> Replacement unit 12 of CPU1
Inserts the name of the variable semantic feature into the portion of the fixed sentence stored in the working memory 5 corresponding to the variable semantic feature not replaced in step A10 in the Japanese sentence / English sentence. In that form, it is stored in the working memory 5, and the process proceeds to step A11.

Now, in the sentence example 5, “Suzuki” corresponding to the “person name” variable semantic feature in the Japanese sentence and “Mr. Su” in the English sentence
Replace "zuki" with the name "person name" of the variable semantic feature. In fact:

[0175] The data is stored in the working memory 5 in this manner, and the process proceeds to Step A11.

Subsequent steps are the same as in the basic embodiment. In steps A11 to A15, the sentence example 5 is displayed on the liquid crystal display 14 as shown in FIG. 13, and the series of translation operations ends.

<Embodiment of Claim 6> An embodiment of the machine translation apparatus according to claim 6 will be described below. In this claim, the second language is honored for the personal name (Mr./Ms., Etc.)
In the case where it is necessary to attach a title and it is not necessary to distinguish it in the first language, it is necessary to prepare a mechanism that can give a correct title in the first language. The case of realizing with semantic features will be described as an example. The “honorific” variable semantic feature is defined in the variable semantic feature dictionary 9 as follows.

The process of translating the input sentence of male Mr. female Ms. input sentence is the same as that of the basic embodiment except for step A8. Therefore, only the difference in step A8 will be described in detail here. Since the operation flow is the same as that of the basic embodiment, it is the same as the flowchart of FIG. Now, it is assumed that the user inputs “Can you call Mr. Suzuki?”, And the example sentence 7 in the bilingual database 7 is searched.

[Bilingual Database 7] Sentence 7: Could you call Mr. Suzuki (male)? Would you page ｛Mr.｝ ｛Suzuki｝? ｛Person's name｝ ｛Title｝ Can you call me? In this claim, in step A8 of the basic embodiment, if there is “variable semantic feature in the fixed form, enter Even if the sentence does not have a corresponding semantic feature, the score will not be reduced. "

Now, in sentence example 7 and the input sentence, “person name”
The four semantic features of "san", "call" and "would you give me" match. Only the "honorific" variable semantic feature in the fixed phrase does not match, but the score is not lowered for this. Therefore, the number of coincidences 4 is multiplied by 100, and the degree of coincidence is set to 400 points.

The subsequent steps are the same as in the basic embodiment. In steps A9 to A15, the sentence example 7 is the liquid crystal display 1
4 is displayed as shown in FIG. 14, and a series of translation operations is completed.

Further, when the user can replace the variable semantic feature, the user can replace the underlined variable semantic feature element from the state of the translation result display of FIG.
If "Suzuki-san" is a woman, the English sentence "Mr."
s. ", and get a translated sentence that matches the situation.

[0183]

【The invention's effect】

<Effect of Claim 1> In the invention according to claim 1, the user can input by correctly grasping a state where sentence constituent elements other than the variable semantic features almost match and the variable semantic features do not match. The fixed sentence whose variable semantic features do not match but are semantically close to the sentence can be reliably displayed.

<Effect of Claim 2> In the invention according to claim 2, a character string that is different in character type from other parts in the input sentence and is not registered as a semantic feature is regarded as one unknown word. Accordingly, it is possible to suppress the display of a sentence having a meaning far from the sentence input by the user, and to surely display a sentence having a close meaning.

<Effect of Claim 3> In the invention according to claim 3, a character string which is different in character type from other parts in the input sentence and is not registered as a semantic feature is regarded as an arbitrary variable semantic feature. Thus, even when the user inputs a word that is not registered as a variable semantic feature element such as a proper noun, a sentence with a similar meaning can be reliably displayed.

<Effect of Claim 4> In the invention according to claim 4, a character string that is different in character type from other parts in the input sentence and is not registered as a semantic feature is used in a second language. By converting it to a species and displaying it as a translation, even if the user enters a proper noun in the first language,
Can be translated in a second language.

<Effect of Claim 5> In the invention according to claim 5, when a variable semantic feature does not match, a specific character or character string is inserted and displayed, so that the user can change the variable meaning. It is obvious that the features did not match, and the visibility of the matching of other components of the sentence is improved. In addition, there is an advantage that it is possible to grasp what elements can be included in the variable semantic feature.

<Effect of Claim 6> In the invention according to claim 6, since the essential expression in the second language exists as a variable semantic feature in the first language, the user subsequently selects the correct element. By doing so, it is possible to always generate an expression that follows the expression in the second language.

[Brief description of the drawings]

FIG. 1 is a block diagram of an embodiment of a machine translation apparatus according to the present invention.

FIG. 2 is an external view of a machine translation apparatus according to an embodiment of the present invention.

FIG. 3 is a flowchart of an embodiment of a translation process according to the present invention.

FIG. 4 is an example of a display screen in one embodiment of the translation processing of the present invention.

FIG. 5 is an example of a display screen in one embodiment of the translation processing of the present invention.

FIG. 6 is an example of a display screen in one embodiment of the translation processing of the present invention.

FIG. 7 is an example of a display screen in one embodiment of the translation processing of the present invention.

FIG. 8 is an example of a display screen in one embodiment of the translation processing of the present invention.

FIG. 9 is an example of a display screen in one embodiment of the translation processing of the present invention.

FIG. 10 is a flowchart of an embodiment of the present invention.

FIG. 11 is an example of a display screen in one embodiment of the translation processing of the present invention.

FIG. 12 is an example of a display screen in one embodiment of the translation processing of the present invention.

FIG. 13 is an example of a display screen in one embodiment of the translation processing of the present invention.

FIG. 14 is an example of a display screen in one embodiment of the translation processing of the present invention.

FIG. 15 is an example of a display screen in one embodiment of the translation processing of the present invention.

FIG. 16 is an example of a display screen in one embodiment of the translation processing of the present invention.

FIG. 17 is a flowchart of an embodiment of the present invention.

FIG. 18 is a flowchart according to an embodiment of the present invention.

FIG. 19 is a flowchart of an embodiment of the present invention.

FIG. 20 is a flowchart of an embodiment of the present invention.

FIG. 21 is a flowchart of an embodiment of the present invention.

[Explanation of symbols]

 Reference Signs List 1 CPU 2 liquid crystal display 3 display driver 4 input recognition unit 5 working memory 6 ROM 7 bilingual database 8 semantic feature dictionary 9 variable semantic feature dictionary 10 analysis unit 11 search unit 12 replacement unit 13 control unit 14 liquid crystal display 15 dedicated pen 16 Handwriting recognition board 17 interpreter icon 18 bus

Claims (6)

[Claims] 1. A bilingual database in which an example sentence of a conversation in a first language and a sentence in a second language, which is a translation thereof, are registered as a pair, and a semantic feature and a notation to be extracted as the semantic feature are defined. A semantic feature dictionary stored as such, a variable semantic feature dictionary having the same semantic feature and replaceable words registered therein, input means for inputting an input sentence for searching the bilingual database, and the input Search means for searching the bilingual database, the semantic feature dictionary and the variable semantic feature dictionary based on the input sentence input from the means, and an example sentence searched from the bilingual database and the semantic feature dictionary by the search means. On the other hand, a replacement unit that performs replacement based on the variable semantic feature dictionary, an example sentence extracted by the search unit and the replacement unit,
An analyzing unit that calculates a degree of coincidence with the input sentence input from the input unit; and an output unit that outputs a translation of an example sentence based on the analyzing unit. If there is a variable semantic feature in the retrieved translated sentence, there is no feature corresponding to the variable semantic feature in the input of the input means, and if there is only one semantic feature that does not match, the variable semantic feature is matched. A machine translation apparatus characterized in that a degree of coincidence is obtained as a result of the translation. 2. The analysis means according to claim 1, wherein the input sentence is a partial character string having a character type different from that of another part and not registered as a semantic feature in the semantic feature dictionary and the variable semantic feature dictionary. The machine translation apparatus according to claim 1, wherein the machine translation apparatus recognizes the word as a word and obtains a degree of coincidence. 3. The analysis means according to claim 1, wherein the input sentence is a partial character string having a character type different from that of another part and not registered as a semantic feature in the semantic feature dictionary and the variable semantic feature dictionary. 2. The machine translation apparatus according to claim 1, wherein the apparatus is recognized as a specific variable semantic feature and obtains a degree of coincidence. 4. The substitution means according to claim 1, wherein the input sentence is a partial character string having a character type different from that of another part and not registered as a semantic feature in the semantic feature dictionary and the variable semantic feature dictionary. 2. The machine translation apparatus according to claim 1, wherein the machine translation is performed to a specific character type. 5. When a character string corresponding to a variable semantic feature in the retrieved translated sentence does not exist in the input sentence, the replacement means inserts and displays a specific character string in the variable semantic feature. 2. The machine translation apparatus according to claim 1, wherein the translation is performed. 6. The replacement means, wherein a linguistic expression that is always expressed in the second language but is not necessarily expressed in the first language is converted into a variable semantic feature in the second language and assigned to a translation result. 2. The machine translation apparatus according to claim 1, wherein the information is displayed.