JPH01214963A - Device for consulting dictionary - Google Patents

Abstract

PURPOSE:To reduce the burden of users by collectively sending a document to a system and finding and presenting a word to be found from a dictionary in accordance with a discrimination reference previously designated by the users. CONSTITUTION:When a user inputs a text, a file editing section 15 receives the text through a user input controller 12 and dictionary consultation controlling circuit 11 and prepare a document file which is stored in a document memory 16. Then a word list preparing section 21 reads out a document file which is an object of dictionary consultation. After disintegrating the document into word strings, the section 21 retrieves dictionary data 18 by using the words as keys and checks the importance of the words. Then the section 21 prepare a word position table 22 and dictionary consulted result table 23 in accordance with importance. A document outputting section 25 reads out the document file from the document memory 16 and outputs the results of dictionary consultation performed in accordance with the contents of the tables 22 and 23 together with the original document to a printer 26 for printing. Thus, the burden of users is reduced.

Description

Detailed Description of the Invention: "Industrial Application Field" The present invention relates to information processing devices such as combi coaters and word processors, and particularly relates to information processing apparatuses such as combi coaters and word processors, and in particular to information processing apparatuses that use electronic dictionaries to find out the meanings of specific words in document files. The present invention relates to a device that adds and presents a document file to an original document file.

``Prior Art'' As the storage capacity of convenience stores and word processors increases, the contents of existing dictionaries are being digitized and utilized. In fact, there are several patent applications related to systems that handle these electronic dictionaries.
No. 30194 describes a system for inputting a keyword and interactively searching for content corresponding to the keyword.

``Problems to be Solved by the Invention'' By the way, in the conventional electronic dictionary system described above, each time a word of interest was encountered in a document, the word was sent to the system and looked up in the dictionary. In other words, from the point of view of the user, it was simply an electronic version of an existing dictionary to make it easier to search, and it did not go beyond the usage of existing dictionaries.

Therefore, an object of the present invention is to send documents all at once to a system, and to find words to be looked up in a dictionary based on criteria specified by the user in advance for determining whether or not to instruct a dictionary lookup. The object of the present invention is to provide a means for presenting information in a form that is easy for the user to understand, such as by mixing it with the original document or adding it separately.

"Means for Solving the Problem" The present invention provides (i) an input means for electronically inputting a sentence to be processed; (ii) a dictionary capable of electronically searching for the meaning of each word; (iii) a search condition input means for inputting conditions for selecting words to be searched from the text input by the input means; and (iv)
(v) extracting means for extracting the meanings of each word selected by the search condition input means using a dictionary; and (v) extracting the meanings of each word selected by the search condition input means, and The dictionary lookup device is equipped with a storage means for storing data in correspondence with each other.

If each word whose meaning has been extracted by the dictionary in the sentence stored by the storage means is marked with a marking to distinguish it from other parts, it is easy to see at a glance which words have annotations prepared for them. and convenient.

In addition, for sentences stored by the storage means, the meanings of words are displayed as notes near each word whose meaning has been extracted by the dictionary in the sentence input by the input means. It is convenient in that the meaning can be understood immediately.

For each word whose meaning has been extracted by the dictionary in a sentence input by the input means, each word meaning may be displayed as an annotation in the vicinity of the word in this way, but the dictionary lookup result It may be stored and displayed as .

In addition to the meaning of each word in the dictionary, a keyword may be added to select the word to be searched, and a separate list of word groups matching the conditions for selecting words may be added. It may be prepared.

In this way, the present invention allows the user to enter in advance general conditions for selecting the words he or she wants to search, so there is no need to search for each word one by one.
It can be adapted according to the user's wishes.

"Example" Hereinafter, the present invention will be described in detail with reference to Examples.

FIG. 1 is a block diagram showing the configuration when the dictionary lookup device of the present invention is applied to a bilingual word processor in Japanese and English. First, each component in the figure will be briefly explained. The dictionary lookup control circuit 11 is composed of a microprocessor and its peripheral circuits, and performs overall control of the word processor. When a user inputs data or commands by operating the keyboard 13 or mouse 14, the user input control section 12 receives the input and sends it to the dictionary lookup control circuit 11. The file editing section 15 receives input data sent from the user input control section 12 through the dictionary lookup control circuit 11, and creates and edits a document file. Document memory 16
is a semiconductor memory for storing document files created by the file editing section 15. The CRT display 17 displays document files and other command instruction menus according to instructions from the dictionary lookup control circuit 11. The dictionary data 18 (electronic dictionary) is a hard disk device containing a plurality of dictionary units for each type of dictionary. The word list generation unit 21 reads out the document file to be searched in the dictionary from the document memory 16, and converts it into -
After decomposing into odd car word strings, each word is looked up in the dictionary while referring to the dictionary data 18, and the attributes of the word, such as importance, type of part of speech, category, etc. are checked, and based on this, the word position table 22 is created. A dictionary lookup result table 23 is created. The word position table 22 is a table that manages each word in a document as a set of the number of characters in that word and an identification code indicating whether or not to perform a dictionary lookup. The dictionary lookup result table 23 is a table that manages the words registered in the word position table 22 that are to be looked up in the dictionary, as a set of the index (usually the name of the word), identification code, and its explanatory text. be.

The dictionary search unit 24 searches the dictionary data 18 based on the headings in the dictionary lookup result table 23, and fills in the explanatory text items in the dictionary lookup result table 23 with the obtained dictionary lookup results. The document output unit 25 is a document memo IJ l
6, the document file is looked up in a dictionary based on the contents of the word position table 22 and the dictionary lookup result table 23, and the result is printed out to the printer 26 along with the original document.

Next, the operation of the Japanese-English bilingual word processor will be explained based on this diagram. In the following explanation, an English document file is read, the most important English words are looked up in a dictionary, and a Japanese explanatory text as a result of the dictionary lookup is added to the original English document file and then printed out on the printer 26. This will be explained based on an example. First, the user uses the keyboard 13 or mouse 14 to input English sentences. Of course, other input methods can also be used. Such methods include (i) a method of reading and inputting a document already stored on a floppy disk, and (11)
) For example, there is a method of using e-mail to use documents sent from overseas.

By using a document that has already been created in this way, the text can be processed more quickly than when the text is directly input, so the effect of dictionary lookup according to the present invention is increased accordingly.

Now, when the user inputs an English sentence using the keyboard 13 or the mouse 14, the file editing section 15 receives this sentence through the user input control device 12 and the dictionary lookup control circuit 11, and then converts it into an English document. A file is created and stored in the document memory 16. Subsequently, the dictionary lookup control circuit 11 moves to dictionary lookup printing processing based on the English document file stored in the document memory 16.

FIG. 2 shows the overall flow of dictionary lookup printing processing in a Japanese/English bilingual word processor. When the user specifies a document file to be processed by word processing, a display as shown in FIG. 3 appears. At this time, a command menu A- is displayed together with the contents of the specified document file (step 2 in FIG. 2). In addition to the dictionary print command, the command menu includes commands such as print, save, edit, and end.

However, since the details of these processes are not essential to the present invention, they are not shown in FIG.

Below is the English sentence “7hisarti” shown on the display screen.
cle presents techniques f
or managing very large
wledge base to supportmul
Tiple 1nference mechanism
The dictionary lookup process will be explained using an example of processing ``s for logicprogramming,''. Also, in this example, in the above English sentence, ``article''
“techniques”, “1nference”
” is a word that you want to look up in a dictionary.

When performing dictionary lookup printing (step 2), the user moves the mouse 14 to move the mouse pointer to the "Dictionary lookup print" position on the command menu and selects it by pressing the mouse button.

If a process other than this is to be performed (step ■), the other process (step ■) can be executed by moving the mouse pointer to the corresponding position on the command menu and selecting another command.

When you select “Dictionary Print” from the command menu,
The dictionary lookup control circuit 11 then controls the CRT display 17.
Display the option menu above (step ■) and request selection of an option. There are two options: specifying the dictionary to be used and specifying search conditions.

FIG. 4 is a display example of the option menu.

The user inputs the dictionary name and search conditions he/she wishes to use from the keyboard 13 and specifies options. Dictionaries used include an English-Japanese dictionary, a Japanese-English dictionary, a traditional Japanese dictionary, a traditional Japanese dictionary, a German-Japanese dictionary, a Russian-Japanese dictionary, an Indo-Japanese dictionary, a Dutch-Japanese dictionary, a mathematics dictionary, an information processing terminology dictionary, a medical terminology dictionary, and an electronic engineering terminology dictionary. etc. can be used. In addition, search conditions can be specified by importance, part of speech, category, or a combination of these. In this example, an English-Japanese dictionary is selected as the dictionary to be used. When the name of the dictionary to be used is specified, the dictionary lookup control circuit 11 issues a channel setting command to the dictionary data 18 to open a channel to the English-Japanese dictionary (step 2). In this example, the search condition is “important eccentric”.
Select (step ■). Thereafter, the dictionary lookup control circuit 11 transfers control to the word list generation section 21. The word list generation unit 21 reads an English document file to be looked up in a dictionary from the document memory 16, decomposes it into a -variable word string, searches the dictionary data 18 using this word name as a key, and searches the dictionary data 18 for the word. The importance level is checked and based on this, a word position table 22 and a dictionary lookup result table 23 are created. At this time, information indicating that dictionary lookup is necessary is recorded in word position table 22 and dictionary lookup result table 23 for words with high importance. In this example, “art”
icle-1 “techniques”, =inf
erence" is decided to be looked up in the dictionary.

FIG.
2 is a flowchart showing processing when generating a dictionary lookup result table 23. First, the word length counter, space length counter, and identification code are reset by initialization processing (step ①).

Next, the process moves to reading the English document file in the document memory 16. It is checked whether the English document file is at the file end or not. If it is the file end (step 2), the process is terminated and control is returned to the dictionary lookup control circuit (step 2).

If it is not the end of the file (step ■), read the characters until the next word appears in the text, add the number of read characters to the space length counter (step ■), and store the read word in a word variable (step ■). , set the length of the read word in the word length counter (step ■). Next, return the read word to its original form (step ■). However, this part of the process is complicated and there are several ways to implement it, so this will be explained in detail in the supplementary explanation section. Next, the dictionary data 18 is searched using the original form of this word as a key, and the attribute part is extracted from the record regarding this word (step ■).
.

FIG. 6 is a conceptual diagram showing an example of the structure of a dictionary.

As shown in the figure, the dictionary is configured as a record consisting of fields of a heading part 31, an attribute part 32, and an explanation part 33. Of these, the attribute i32 is further divided into subfields of a part of speech field 34, a category field 35, and an importance field 36. Become. The operation of looking up a dictionary is to give a word and search among these records for a record in which the given word name and the contents of its heading section 31 match.

The importance level is an integer from 0 to 3, and the larger the value, the more important it is. Therefore, for example, when "important weirdo" is specified as a search condition, a value of 3 in the importance field 36 is used as a dictionary lookup condition. Search conditions can specify not only importance, but also parts of speech, category types, or a combination of these.

In this embodiment, the search condition (
In this case, “Important Stranger”), and if the search conditions are met (step ■), set the space length counter value in the space length field 41 of the word position table 22, and set the value zero to the identification code field 42. Put each (
Step ■), reset the value of the space length counter.Step ■), increment the value of the identification code (step o), and set the value of the word length counter in the space length field 41 of the word position table 22, and set the value of the word length counter in the identification code field 42. Enter the value of the identification code into each (step ■)
. In addition, the value of the identification code is entered in the identification code field 43 of the dictionary lookup result table 23, the original form of the read word is entered in the heading field 44 (step ■), and the value of the word length counter is reset (step ■). The process moves on to reading the next word in the text (step O). Also, if the previous search condition is not satisfied (step @), add the word length counter value to the space length counter value (step ■), reset the word length counter value, then proceed to the next step). The process moves on to reading the next word inside (step ■). That is, in this case, the word position table 22 and dictionary lookup result table 23 are not updated.

Figure 7 shows how the English sentence displayed in Figure 3 is generated by the word list generator 2.
1 shows a part of the words broken down into word strings.

Figure 8 shows word list generation a' based on the word strings in Figure 7.
[This shows the structure of the word position table generated by s21. In the figure, the value of the identification code field 42
0'' is the default value and indicates that dictionary lookup is not performed for the corresponding word, and other values indicate that dictionary lookup is performed for the corresponding word.

Further, FIG. 9 shows the structure of the dictionary lookup result table 22 generated by the word list generation section 21 based on the word strings shown in FIG. 7. In the figure, the explanation field 45 is completely blank at this point.

'After this, the dictionary lookup control circuit 11 transfers its control to the dictionary search section 24. The dictionary search section 24 retrieves an English word from the heading field 44 of the dictionary lookup result table 22, performs a dictionary lookup while referring to the dictionary data 18 using this word as a key, reads the contents of the explanation section of the dictionary lookup result, and searches the dictionary for the explanation field. Write the explanation of the English word in 45 (Second
Figure step ■).

FIG. 1O is a flowchart showing a process in which the dictionary search unit 24 performs a dictionary lookup and embeds an explanatory sentence of an English word in the explanation field 45 of the dictionary lookup result table 23.

First, while reading the dictionary lookup result table 23, it is determined whether it is a table end or not, and if it is not a table end (step ■), the pointer pointing to each record in the dictionary lookup result table is updated (step ■), and the record pointed to by this pointer is updated. Read an English word from the heading field and put it in a word variable (step ■), and use this word variable as a key to look up this English word in a dictionary while referring to dictionary data 18 (step ■). After writing the explanation of the English word among the dictionary lookup results into the explanation field 45 of this record (step ■), the process moves on to the next record in the dictionary lookup result table 23 (step ■).

If a table end is detected in the previous table end determination (step 2), the dictionary search unit 24 ends the process and returns control to the dictionary lookup control circuit (step 2).

FIG. 11 shows the contents of dictionary lookup result tables 2 and 3 generated by the dictionary search unit 25 based on the dictionary lookup result table and dictionary data 18 of FIG. In the figure, the explanation field 53 is filled with an explanation of the corresponding word.

Thereafter, the dictionary lookup control circuit 11 transfers its control to the document output section 25. The document output unit 25 refers to the word position table 22, the dictionary lookup result table 23, and the document memory 16, and draws an underline under the dictionary lookup word in the original English sentence, and also converts the dictionary lookup result into the original English. is added to the end of the sentence and printed out on the printer 26 (step ■).

FIG. 12 is a flowchart showing document output processing performed by the document output unit 25. First, l records are read from the word position table 22 (step ■), character strings corresponding to the number of characters shown in the space length field 41 are read from the original English document file (step ■), and character strings corresponding to the number of characters shown in the space length field 41 are read from the original English document file (step ■). Check the contents (step ■), and if this is the default value (that is, the value is zero) (step ■), the read character string is output as is to the printer 26 without underlining. In other words, if the value is other than zero (step ■), the read character string is underlined and output (step ■). Next, it is checked whether the word position table is at the table end or not, and if it is not the table end, the process moves on to the next record (step 2), and if it is the table end, the process moves on to outputting the subsequent dictionary lookup result (step 2). At this time, first perform a page break (step 0),
Read one record from the dictionary lookup result table 23 (step ■), and output the contents of the heading field 52 and the explanation field 53, respectively (step ■)
, check whether the dictionary lookup result table is a table end or not, and if it is not a table end, move on to processing the next record (step @), and if it is a table end (step ■
) The document output unit 25 finishes the processing and outputs the dictionary lookup control circuit 1.
The control is returned to step 1 (step ■).

As a result, the user can obtain an underlined English sentence and a dictionary lookup result for the English word looked up in the dictionary from the printer 26.

FIG. 13 shows an example of the text output by the document output unit 25 at this time. Thereafter, the dictionary lookup control circuit 11 moves to display a menu screen (step 0) and processes another document file.

Next, the dictionary lookup device described above will be explained in more detail.

First, among the processes performed by the word list generation unit 21, a method for deriving the original form from the read word will be described.

In this case, if all words other than proper nouns used in English, including their inflections, are registered in the dictionary, such a word transformation algorithm is not necessary. However, English words have inflections, and even the same word can change in various ways depending on the context. If all words, including these inflections, were registered in a dictionary, the dictionary would be extremely large.

Therefore, dictionaries usually use only the original form of the word as a heading, excluding the suffix that follows the word through regular changes. Therefore, when looking up a dictionary, it is necessary to remove this suffix. In this embodiment, words that undergo regular inflection such as past tense, plural tense, progressive tense, etc. are returned to their original forms.

Words that change irregularly are treated as separate words.

Next we show an algorithm to do this.

FIG. 14 is a flowchart showing the basic part of the process of restoring words read from an English document to their original form.

First, when reading words from an English document (step ■),
Remove non-English characters such as spaces, periods, and commas from the end of the read word (step ■). Next, a dictionary lookup is performed using the resulting character string as a key, and if this is found in the dictionary data 18 (step ■), the dictionary lookup result is returned as the original form of the word (
Step ■). If the corresponding word is not found as a result of this dictionary lookup (step ①), the original form is further searched for in the next step. First, it is checked whether this character string ends with the alphabetic character °s°, and if it ends with the alphabetic character "So" (step ■), the plural and third person singular forms are returned to their original forms (step ■ ).

FIG. 15 is a flowchart showing the process of returning plural and third person singular forms to their original forms. After removing the English character “So” from the end of the given character string (step ■) Return the dictionary lookup result as the original form of the word (step ■).

Also, if the corresponding single 35 is not found as a result of this dictionary lookup (step ■), after removing the English character 1 el from the end of the given character string (step ■),
A dictionary lookup is performed using the obtained character string as a key, and if it is found in the dictionary data 18 (step ■), the dictionary lookup result is returned as the original form of the word (step ■).
. Also, if the corresponding word is not found as a result of this dictionary lookup, step ■) Furthermore, after rewriting the last character of the given character string from "Io" to "y" (step ■), perform the dictionary lookup again. , if this is found in the dictionary data 18 (step O), the dictionary look-up result is returned as the original form of the word (step ■).As a result, for example, "hats', "CatCheS ''', "1
“Adies” etc. are each returned to their original form and “hat” etc.
”, “catch”, and “1ady”.

Also, if the corresponding word is not found as a result of this dictionary lookup (Step @), return as is (Step 0)
).

Next, in the processing of the basic part above, the character string is an English character ゛S°
If it does not end with the alphabetic character "do", check whether this character string ends with the alphabetic character ``do'' (Step ■), then return the past tense to its original form. (Step 0).

FIG. 16 is a flowchart showing the process of restoring the past tense to its original form. At this time, the alphabetic character “d” starts from the end of the given string.
After o is removed (step ■), a dictionary lookup is performed using the obtained character string as a key, and if this is found in the dictionary data 18 (step ■), the dictionary lookup result is returned as the original form of the word ( Step ■). If the corresponding word is not found as a result of this dictionary lookup (step ①), then the English letter "e" is added from the end of the given character string.

After removing (Step ■), a dictionary lookup is performed using the obtained character string as a key, and if this is found in the dictionary data 18 (Step ■), the dictionary lookup result is returned as the original form of the word (Step ■). Also, if the corresponding word is not found as a result of this dictionary lookup (step ■), the last character of the given character string is added from l° to “
After rewriting it to y' (step ■), the dictionary is looked up again, and if this is found in the dictionary data 18 (step ■), the dictionary lookup result is returned as the original form of the word (
Step ■). Also, if the corresponding word is not found as a result of this dictionary lookup (step 0), if the same character continues from the end of the given character string, remove this character (step ■), and then the obtained character A dictionary lookup is performed using the column as a key, and if it is found in the dictionary data 18 (step 2), the dictionary lookup result is returned as the original form of the word (step 2). This allows for example “hoped”5=picked”, “Cr1e
d”.

'hopped' etc. are returned to their original form and 'h
``ope'', ``pick'', ``cry'', ``hop''
”.Furthermore, if the corresponding word is not found as a result of this dictionary lookup, the process returns as is (step ■).

Next, in the processing of the basic part above, the string is changed to the English character “do”.
If it does not end with the character string "ing" (step ■), it is further checked whether this character string ends with the character string "ing". If the string ends with “Ing” (step ■)
performs processing to return the progressive form to its original form (step 0).

FIG. 17 is a flowchart showing the process of restoring the progressive form to its original form. At this time, the character string “i” starts from the end of the given character string.
After removing ``ng'' and adding English character 1 el (step ■), look up the dictionary using the obtained string as a key,
If this word is found in the dictionary data 18 (step ■), the dictionary lookup result is returned as the original form of the word (step ■).

If the corresponding word is not found as a result of this dictionary lookup (step ■), remove the English character "eo" from the end of the given string (step ■), and then use the obtained string as a key. A dictionary lookup is performed, and if it is found in the dictionary data 18 (step ■), the dictionary lookup result is returned as the original form of the word (step -).Also, if the corresponding word is not found as a result of this dictionary lookup, Step ■) Furthermore, when the same character continues from the end of the given character string, after removing this character (
Step ■), perform a dictionary lookup using the obtained character string as a key, and if this is found in the dictionary data 18 (
Step 0) Return the dictionary lookup result as the original form of the word (Step ■).

This allows for example "hopping", "jumping"
ing”, “hopp ing”, etc. are returned to their original forms and become “hope”, “jump”, “
hop''. If the corresponding word is not found as a result of this dictionary lookup (step 2), the process returns directly (step 0).

Next, in the process of the basic part shown in Figure 14, the process of returning the plural/third person singular to its original form, the process of returning the past tense to its original form, and the process of returning the progressive tense to its original form all involve dictionary lookup. The case where the corresponding word is not found will be explained (Step 2 in FIG. 14). In this case, it is concluded that the word does not exist in the dictionary, e.g. some kind of proper noun (step ■).
, the process ends (step @).

The fact that it did not exist in the dictionary either in its original form or in its original form means that it is either a proper noun, a very rare word that is not in the dictionary, or a word that is misspelled. It is thought that it is one of the following.

The dictionary lookup device of this embodiment does not perform any processing on these words, but if the fact that the word does not exist in the dictionary itself is considered important, such a word may be underlined. It is also effective to retrieve the search term and restore it as an unsearchable word.

Now, in the example of the algorithm that returns words to their original forms as explained above, in order to simplify the explanation, we have focused on words that undergo regular inflection such as past tense, plural tense, and progressive tense. In fact, in addition to such regular inflections, it is also possible to deal with inflections caused by English prefixes as shown in Table 1 below and suffixes as shown in Table 2.

(Space below) Table 1 (English prefixes) (Space below) Table 2 (English suffixes) "First variant" In the example described above, the dictionary lookup results are summarized at the end of the document. presentation. In the first modified example, an example will be briefly described in which the result of dictionary lookup is presented immediately below the word underlined by dictionary lookup in the original English document.

FIG. 18 is an example of the output results presented by this method. In this case, the word list generation section 21 and the dictionary search section 2
The processing in step 2 is the same as that in the embodiment described above. However, the processing of the document output unit 25 is slightly different. Next, this part will be explained.

FIG. 19 is a flowchart showing the processing of the document output unit 25 when the dictionary lookup result of an underlined word in the original document is presented immediately below the word.

First, a character string is read from the word position table 22 (step ■), a character string corresponding to the number of characters shown in the space length field 41 is read from the original English document file (step ■), and a character string corresponding to the number of characters shown in the space length field 41 is read (step ■).
Examine the contents of (step ■) and confirm that this is the default value (
That is, if the value is zero (step ■), the read character string is output as is without underlining (step ■). If this is not the default value (that is, a value other than zero) (step ■), the read character string is underlined and output (step ■), the display line is moved down one line (step ■), and the dictionary is looked up. The result table is read as a code (step ■), and after outputting the contents of the heading field, yield 52, and explanation field 53, respectively, the display line is returned to its original state (step 0). Next, it is checked whether the word position table is a table end or not, and if it is not a table end, the process moves on to the next record (step ■), and if it is a table end (step 0), the document output unit 25 ends the process. Then, the control is returned to the dictionary lookup control circuit 11 (step 2). Note that in the above example, the words found in the dictionary were underlined, but it would also be effective to do things like shading, coloring, changing the font size, or changing the font design. Alternatively, you can number the words you look up in a dictionary.

Another example of presenting dictionary lookup results is to create a vertically long area next to the presentation area of the original document and present the dictionary lookup results at the position corresponding to the line where the underlined word exists. It is.

FIG. 20 is an example of the output results presented by this method. In this case as well, the processing of the word list generation unit 21 and the dictionary search unit 24 is the same as in the embodiment described above. This is achieved by slightly changing the processing of the document output section 25, but the details are not that different from the flowchart explained above, so the explanation thereof will be omitted.

"Second Modification" Generally, data on memory can be accessed at high speed, but its storage capacity is small. Conversely, data on an external storage device such as a hard disk drive takes time to access, but its storage capacity is extremely large. In the example, the dictionary was configured by a hard disk device, but
Practical speed cannot be achieved when the number of dictionary lookups is large. Therefore, several modifications can be made, particularly to the dictionary search section. If these modifications are used in combination, even greater effects can be expected.

(i) Preparation of dictionary for original form conversion.

The functions of a dictionary for restoring each word to its original form and a dictionary for extracting an explanatory sentence (the meaning of a word) are different. Therefore, it is effective to use separate dictionaries for these. In this case, by arranging the former dictionary in memory, it is possible to speed up dictionary lookup.

(ii) Preparation of a dictionary of words that do not need to be searched.

There are words such as "we", "is#", "to", etc. that are common sense to people who read English and do not need to be looked up in a dictionary. By searching only for words that are not in the dictionary, there is no need for wasteful processing of looking at the attributes of the dictionary and applying conditions, making it possible to speed up dictionary lookups.

(iii) Dividing the dictionary by purpose.

Even in the same dictionary, it is necessary to search the attribute section for all words in order to select words that require explanations. On the other hand, the search in the commentary section only needs to be performed for the selected words. Therefore, by placing a dictionary containing only headings and attribute sections in memory, and placing a complete dictionary with an explanation section on the hard disk, it is possible to speed up dictionary lookup.

(iv　Dictionary of dictionary by frequency.

Split the dictionary into words with high frequency and words with low frequency. The former is placed in memory, and the latter is placed on the hard disk. This makes it possible to speed up dictionary lookup.

"Effects of the Invention" As explained above, in the dictionary lookup device of the present invention, in addition to information on each word registered in the electronic dictionary, attribute information such as importance, type of part of speech, category, etc. Included. Then, the entire document is accepted at once, and each word is looked up in the dictionary one by one.The attribute section is first checked, and the dictionary lookup results are presented only if the content matches the search conditions specified by the user in advance. I decided to do so. This allows the system to look up the meaning of words that the user wants to look up in the dictionary in advance, dramatically increasing the efficiency of using electronic dictionaries, especially when reading foreign language literature. It has the effect of aiding understanding.

[Brief explanation of the drawing]

1 to 17 are for explaining one embodiment of the present invention. FIG. 1 is a block diagram showing the configuration of a bilingual word processor in Japanese and English, and FIG. 2 is a block diagram showing the entire dictionary lookup printing process. A flowchart showing the flow, Figure 3 is a diagram showing an example of a screen displaying a command menu, Figure 4 is a diagram showing an example of a screen displaying an option menu, and Figure 5 is a diagram showing the processing of the word list generation section. Figure 6 is a conceptual diagram showing an example of the dictionary structure, Figure 7 is a conceptual diagram showing the state when an English document is broken down into word strings, and Figure 8 is the structure of a word position table. 9 is a conceptual diagram showing an example of the structure of the dictionary lookup result table in a state where the explanation field is not filled in. FIG. 10 is a conceptual diagram showing the processing of the dictionary search section. Figure 11 is a conceptual diagram showing an example of the structure of a dictionary lookup result table with the explanation field filled in, Figure 12 is a flowchart showing the processing of the document output unit, and Figure 13 is a diagram showing the processing output by the document output unit. A diagram showing an example of text, Figure 14 is a flowchart showing the basic part of the process of returning words read from an English document to their original form,
Figure 15 is a flowchart showing the process of returning plural and third person singular forms to their original forms, Figure 16 is a flowchart showing the process of returning past tenses to their original forms, and Figure 17 is a flowchart of returning progressive forms to their original forms. It is a flow chart showing processing. Furthermore, Figures 18 to 20 are for explaining some modified examples of the present invention, and Figure 18 shows the output when the result of dictionary lookup is presented directly below the underlined word. A diagram showing an example of the result, Figure 19 is a flowchart showing the process when the dictionary lookup result is presented directly below the underlined word, and Figure 20 is a flowchart showing the process when the dictionary lookup result is presented immediately below the underlined word. FIG. 7 is a diagram showing an example of an output result when a dictionary lookup result is presented at a position corresponding to a row. 11... Dictionary lookup control circuit, 12... User input control section, 15... File editing section, 16... Document memory, 17...・・CRT display, 18・・・・
・Dictionary data, 21... Word list generation unit, 22... Word position table, 23... Dictionary lookup result table, 24...
...Dictionary search unit, 25...Document output unit, 26...Printer. Applicant Fuji Xerox Co., Ltd. Agent
Patent Attorney Umeo Yamauchi Figure 2 ■ Figure 3 Figure 5 Character string Figure 7 Data skin Figure 10 Figure 11 (5) Qeni L'' Figure 14 Figure 15 Figure 16 Figure 17 Figure 'Mc18Figure 20

Claims (1)

[Scope of Claims] 1. An input means for electronically inputting a sentence; a dictionary that allows the meaning of each word to be searched electronically; and a word to be searched from among the sentences input by the input means. a search condition input means for inputting conditions for selection; an extraction means for extracting the meanings of each word selected by the search condition input means using the dictionary; and an extraction means for extracting each word meaning extracted by the extraction means. A dictionary look-up device comprising: a storage means for storing words in correspondence with each word in a sentence inputted by the input means. 2. In the sentence stored by the storage means, each word in the sentence input by the input means whose meaning has been extracted by the dictionary is marked with a marking to distinguish it from other parts. A dictionary lookup device according to claim 1, characterized in that: 3. The text stored by the storage means is characterized in that the meanings of the words are annotated near each word whose meaning has been extracted by the dictionary in the text input by the input means. A dictionary lookup device according to claim 1. 4. The dictionary lookup device according to claim 1 or claim 2, wherein each word whose meaning has been extracted by the dictionary in a sentence inputted by the input means is stored as a dictionary lookup result. . 5. The dictionary lookup device according to claim 1, wherein each word in the dictionary has a keyword added thereto in addition to its meaning to select the word to be searched.