JPH11203281A - Electronic dictionary search device and medium storing electronic dictionary search device control program - Google Patents

Abstract

(57) [Summary] [Problem] To perform a dictionary search with a high probability even when a character string of an input sentence is different from a heading of an electronic dictionary due to a difference in a kana, a notation, or the like. An electronic dictionary storing headings and candidates corresponding thereto, an input unit for inputting text and instructing dictionary search, and morphological analysis of the text input by the input unit based on a morphological analysis dictionary. A morphological analysis unit, a heading obtaining unit that obtains a character string morphologically analyzed by the morphological analysis unit as a heading, corrects the utilized word of the obtained heading to a basic form, and stores it in a heading storage unit; and searches the electronic dictionary. A dictionary search unit that converts a heading stored in the heading storage unit into a corresponding candidate; and a display unit that displays an input sentence and the dictionary searched candidate, wherein the character string obtaining unit includes the morpheme All the different notations related to the character string subjected to morphological analysis by the analysis unit are acquired as headings.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention is applied to a word processor, a personal computer, an information terminal or the like provided with an electronic dictionary, and searches words of the electronic dictionary when translating an input sentence displayed on a screen. The present invention relates to an electronic dictionary search device and a medium storing an electronic dictionary search device control program.

[0002]

2. Description of the Related Art Conventionally, for example, when translating a sentence in a native language into a sentence in another language or from a sentence in another language into a sentence in a native language, commercially available general machine translation software is used. Since the translation accuracy of general machine translation software is not very high, the results of machine translation are often unsatisfactory. On the other hand, a personal computer that starts machine translation software can also start a commercially available electronic dictionary (book version dictionary).
Alternatively, when the user wants to check a related translation in an electronic dictionary, the electronic dictionary can be searched for a sentence input at the time of machine translation, but the operability of dictionary lookup is not good.

As a technique for solving this problem, Japanese Patent Laid-Open No.
According to Japanese Patent Application Laid-Open No. 329059, a series of referencing actions performed when a digitized text (original text) written in a certain language is translated into a word dictionary, machine translated, or used in an idiomatic expression. A general-purpose reference device that can be performed by an operation procedure has been proposed.

As a technique for not only extracting target information easily but also extracting relevant information without omission, according to Japanese Patent Laid-Open No. 4-178767, a key word for dictionary search is disclosed. There has been proposed an electronic dictionary search method in which a character string for narrowing down search information is input and a morphological analysis of the character string is performed, and only a character string included in a transliteration part or an example part of the dictionary is displayed.

[0005]

However, in Japanese Patent Application Laid-Open No. H8-329059, for example, when an electronic dictionary is looked up in a dictionary, the dictionary cannot be looked up if the headings are different due to a difference in notation of kana or katakana. Also, in morphological analysis used in machine translation, since the syntax and meaning are not analyzed, there is ambiguity in segment break positions and words. For this reason, dictionary lookup is not possible.

If a morphological analysis dictionary used in machine translation cannot be translated properly unless it is translated in a long phrase, clause, or sentence range, the dictionary is not searchable because the range is specified as a heading as it is. Also, Japanese Patent Application Laid-Open No. 4-1
In Japanese Patent No. 78767, it is necessary to distinguish a search key from an information narrowing character string, and there is a problem that the contents of the electronic dictionary are not configured to be more easily seen.

The present invention has been made in view of the above circumstances. For example, a character string of an input sentence is
Even if it differs from the heading of the electronic dictionary due to differences in notation, etc., by acquiring the character string obtained by morphologically analyzing the sentence and all the different notations related to the character string as the heading of the electronic dictionary, and correcting the utilized words to the basic form An electronic dictionary search device capable of searching an electronic dictionary with a high probability only by inputting a sentence; and a medium electronic dictionary search device storing an electronic dictionary search device control program.

[0008]

According to the present invention, there is provided an electronic dictionary storing a heading and candidates corresponding thereto, an input unit for inputting a sentence and instructing a dictionary search, and a morphological conversion of the sentence input by the input unit. A morphological analysis unit that performs morphological analysis based on the analysis dictionary, and a heading that obtains a character string morphologically analyzed by the morphological analysis unit as a heading, and corrects the utilized word of the obtained heading to a basic form and stores it in a heading storage unit. An acquisition unit, a dictionary search unit that searches the electronic dictionary and converts a headline stored in the headline storage unit into a corresponding candidate, and a display unit that displays an input sentence and the dictionary searched candidate, The character string acquisition unit is an electronic dictionary search device, wherein all of different notations related to a character string morphologically analyzed by the morphological analysis unit are acquired as headings.

In the present invention, an electronic dictionary, a morphological analysis dictionary, a morphological analysis unit, a heading storage unit, a heading acquisition unit, and a dictionary search unit include a CPU, a ROM, a RAM, an I / O
It is composed of a microcomputer consisting of an O port.
In particular, electronic dictionaries and morphological analysis dictionaries are RO
M or EEPROM, floppy disk, CD-
It is composed of a storage device such as a ROM and an MD. The heading storage unit is constituted by a RAM or an EEPROM therein. Input unit is keyboard, pointing device, O
It is composed of CR and the like. The display unit includes an LCD (liquid crystal display) device, a PD (plasma display) device, and the like.

According to the present invention, even if a character string of an input sentence is different from a heading of the electronic dictionary due to a difference in a sentence kana, a notation, etc., a character string obtained by morphologically analyzing the input sentence and a different notation related to the character string. Since all of them are acquired as headings, and the inflected words are corrected to the basic form, it is possible to search the electronic dictionary with a high probability only by inputting sentences. For example, when translating a sentence in one's own language into a sentence in another language, a Japanese-English dictionary can be searched simply by inputting a sentence written in one's own language. You can concentrate on translating into other languages.

[0011] The heading acquisition unit can search the electronic dictionary for a notation that matches a heading stored in the electronic dictionary among different notations related to the character string morphologically analyzed by the morphological analysis unit. May be configured to be added to a dictionary searchable mark and stored in the heading storage unit.

[0012] According to the above configuration, simply by inputting a sentence, a dictionary searchable index is extracted and a dictionary searchable mark is added. Therefore, it is possible to easily check the index and search the electronic dictionary. it can.

If the character string morphologically analyzed by the morphological analysis unit has a plurality of delimiter positions, the heading obtaining unit obtains all of the character strings delimited by the plurality of delimiter positions as headings of the electronic dictionary. Then, the data may be stored in the heading storage unit.

According to the above configuration, even if an input sentence has an ambiguous heading, a character string is cut out at a plurality of delimiter positions and all of the headings are used as headings. Can be.

The apparatus further comprises a compound heading correspondence table in which compound words, each of which is a set of a plurality of headings stored in the electronic dictionary, are stored in advance, and the heading acquiring unit searches the compound heading correspondence table to read the compound heading correspondence table. When the compound word is included in the character string subjected to the morphological analysis according to the above, the heading constituting the compound word may be acquired as the heading of the electronic dictionary and stored in the heading storage unit.
The compound heading correspondence table is stored in the ROM or EE
It is composed of a storage device such as a PROM, a floppy disk, a CD-ROM, and an MD.

According to the above configuration, even if the morphologically analyzed character string includes a compound word that is not in the electronic dictionary, the electronic dictionary can be searched using each word constituting the compound word as a heading.

A sub-key table in which a main key and sub-keys are stored in advance using the headings of the electronic dictionary as primary keys and example phrases thereof as sub-keys, If there is a heading, a sub-key table search unit that searches the sub-key table for a sub-key using a primary key, and an example word that matches the sub-key searched by the sub-key table search unit are displayed at a predetermined position on the screen. A display processing unit for displaying may be further provided.
The secondary key table is stored in ROM or EEPROM.
M, a floppy disk, a CD-ROM, a storage device such as an MD. The sub key table search unit and the display processing unit are configured by a microcomputer including a CPU, a ROM, a RAM, and an I / O port.

According to the above configuration, the sub-keys (example phrases) related to the primary key (headings) are extracted from the input text and the headings and the example phrases of the electronic dictionary are displayed. References can be searched faster.

The apparatus further comprises a sub-key designating unit for designating a sub-key searched by the sub-key table search unit, wherein the display processing unit displays the sub-key designated by the sub-key designating unit at a predetermined position on a screen. The configuration may be such that: In addition,
It is composed of a sub key instruction unit, a keyboard, and a pointing device.

According to the above configuration, since the sub key is designated and displayed at a predetermined position, it is easy to confirm the sub key to be searched in the dictionary.

[0021] The display processing unit may be configured to display a primary key corresponding to a heading of the electronic dictionary and a sub key associated with the primary key in a distinguishable manner. According to the configuration, the primary key and the secondary key are displayed so as to be distinguishable from the other headings, so that it is easy to confirm the headline to be searched in the dictionary.

The display processing section may display a list of sub-headings of the electronic dictionary that match the sub-key. According to the configuration, since the sub key is displayed as possible, it is easy to confirm the sub heading to be searched in the dictionary.

[0023]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in detail based on an embodiment shown in the drawings. The present invention is not limited by this.

FIG. 1 is a block diagram showing the configuration of the electronic dictionary search device of this embodiment. In FIG. 1, reference numeral 1 denotes a pointing device such as a pen and a mouse, a keyboard,
1 shows an input unit including an OCR or the like. Reference numeral 2 denotes a morphological analysis unit that performs morphological analysis of an input sentence based on a morphological analysis dictionary. Reference numeral 3 denotes a heading acquisition unit that acquires, as a heading, a character string morphologically analyzed by the morphological analysis unit 2. Reference numeral 4 denotes a dictionary search unit that searches the electronic dictionary and converts the electronic dictionary into a candidate corresponding to the index acquired by the index acquisition unit 3. 5
Denotes a display processing unit for displaying the headings of the electronic dictionary and example phrases thereof at predetermined positions on the screen. Reference numeral 6 denotes a display unit composed of an LCD (Liquid Crystal Display), a PD (Plasma Display) or the like.

Reference numeral 7 denotes a morphological analysis dictionary including a morphological analysis procedure and a grammar. Reference numeral 8 denotes a morphological analysis result buffer that stores a headline acquired by the headline acquisition unit 3 and also functions as a headline storage unit. Reference numeral 9 denotes a parsing unit for parsing an input sentence based on a parsing dictionary. Reference numeral 10 denotes a parsing dictionary including a procedure for parsing and a grammar. Reference numeral 11 denotes a translation unit that translates the input sentence analyzed by the syntax analysis unit 9 based on the translation dictionary.

Reference numeral 12 denotes a translation dictionary including word data to be translated, procedures, and grammar. 13 is a compound heading correspondence table, 14 is an electronic dictionary storing headings and corresponding candidates, 15 is a sub-key table in which a main key and sub-keys are stored in advance using headings of the electronic dictionary as main keys and example phrases thereof as sub-keys. Is shown. Reference numeral 16 denotes a special cursor key for designating the searched secondary key, and functions as a secondary key designating unit.

The morphological analysis unit 2, heading acquisition unit 3, dictionary search unit 4, display processing unit 5, syntax analysis unit 9, and translation unit 11
It is mainly composed of a microcomputer CPU and a control program. The morphological analysis dictionary 7, the morphological analysis result buffer 8, the translation dictionary 12, the compound heading correspondence table 13, the electronic dictionary 14, and the sub key table 15
M, EEPROM, floppy disk, hard disk, CD-ROM, MD, etc.

In FIG. 1, indicated by a dotted line block,
The syntax analysis unit 9, the syntax analysis dictionary 10, the translation unit 11, and the translation dictionary 12 are configured to execute machine translation.
It does not have to be essential to the present invention. 17 is CD-
Indicates an external storage medium including a ROM, an IC card, an MD, a floppy disk, and the like. The storage medium 17 stores a morphological analysis program, a dictionary search program, a syntax analysis program, a translation program, a display processing program, and various data. I have.

Numeral 18 reads each program and various data stored in the storage medium 17, and reads the morphological analysis unit 2, the dictionary search unit 4, the display processing unit 5, the syntax analysis unit 9, and the translation unit 11.
1 shows a medium reading device to be installed in the system. 19 is CP
U, a ROM, a RAM, and a control unit including a microcomputer including an I / O port.
The entire device is controlled via the bus 20. 20 is the control unit 1
Reference numeral 9 denotes a bus for transferring addresses and various data.

Referring to FIG. 1, according to another aspect of the present invention, an electronic dictionary 1 storing headings and candidates corresponding thereto is provided.
An input unit 1 for inputting a sentence and instructing a dictionary search;
An electronic dictionary search device having a display unit 6 and a computer 1
A medium 17 storing a program controlled by the computer 9, the computer 19 morphologically analyzes the text input by the input unit 1 based on the morphological analysis dictionary 7, and performs a morphologically analyzed character string and its character. All of the different notations related to the columns are acquired as headings, and the conjugated words of the obtained headings are corrected to the basic form and stored in the heading storage unit 8. The electronic dictionary 14 is searched and stored in the heading storage unit 8. A medium 17 is provided that stores an electronic dictionary search device control program that causes the display unit 6 to convert an input sentence and a dictionary searched candidate thereof into a corresponding candidate.

FIG. 19 is a block diagram showing one configuration of the heading acquiring section according to the present embodiment. As shown in FIG. 19, the heading acquisition unit 3 includes a heading storage unit 301 and a heading search unit 3
02, an all-notation acquisition unit 303, an all-notation acquisition unit 304, an all-morpheme analysis result acquisition unit 305, a compound heading correspondence table search unit 306, and a compound heading acquisition unit 307 for acquiring the heading of the compound heading correspondence table 13. ing. In particular, the heading storage unit 301 is a RAM, an EEPROM,
It consists of.

FIG. 20 is a block diagram showing one configuration of the display processing unit according to this embodiment. As shown in FIG. 20, the display processing unit 5 includes a sub-key table search unit 501, an input character string acquisition unit 502, a sub-key table comparison unit 503, a sub-key display processing unit 504, a feature display processing unit 505, and a summary display process. 506.

FIG. 2 is a main flowchart showing the procedure of the translation process / dictionary lookup process according to this embodiment. Referring to FIG. 2, a description will be given of a translation process / dictionary lookup process for translating a Japanese sentence in its own language into an English sentence in another language. Step S1: In the input unit 1, a user inputs a Japanese sentence to be translated into an English sentence. The input method may be any method such as kana-kanji conversion, character recognition, and cut and paste. Step S2: The morphological analysis unit 2 morphologically analyzes the Japanese sentence input in step S1 based on the morphological analysis dictionary / grammar table 7, and cuts out words from the Japanese sentence.
Here, the algorithm of the morphological analysis does not particularly matter.

FIG. 10 is an explanatory diagram showing the procedure of the Japanese-English machine translation process according to this embodiment. As shown in FIG. 10, when a sentence including a clause, a phrase, and a word is input to the morphological analysis unit 2, the morphological analysis unit 2 outputs a character string of a word (including a compound word or the like),
Morphological analysis is performed on the part of speech and the segment breakpoints.

For example, from the input sentence "a certain student went to school", "a", "student", "ga", "school",
The result of the word "he", "go", and "ta" and the part of speech information are obtained. The morphological analysis unit 2 stores the linguistic data in the morphological analysis result buffer 8 for the words with the ending words and the words with the different notations in the morphological analysis result buffer 8 and uses them in the dictionary lookup processing. For example, in the case of the above described notation data “go”, the end form notation data “go” is stored. "Yes", "Student",
You can also get the results of "go to school" and "go".
This phrase information is not required for dictionary lookup processing.

FIG. 11 is a diagram showing an example of different notation data and notation marks for dictionary lookup stored in the morphological analysis result buffer according to this embodiment. As shown in FIG. 11, regarding the different notation data, if “meeting”, “meeting”,
“Meeting”, “meeting”, and “meeting” are stored. “Black circle” shown in FIG. 11 is a notation mark for dictionary lookup, and indicates, among different notation data, different notation data that can be looked up in a dictionary. This mark is added by examining notation data that can be looked up in the electronic dictionary 14 in advance. However, this mark is unnecessary except for the notation data that can be looked up in the electronic dictionary 14, and thus need not be added. That is, only the notation data with the notation mark for dictionary lookup may be stored.

Step S3: Determine whether to perform machine translation or to perform dictionary lookup of the electronic dictionary 14. When performing machine translation, the process proceeds to step S4, and when performing dictionary lookup of the electronic dictionary 14, the process proceeds to step S8. In this embodiment, since the steps S4 and S5 of the machine translation process are not essential,
A brief description will be given. Step S4: The parsing unit 9 parses the input sentence into a tree structure using the parsing dictionary / grammar table 10 from the morphological analysis result obtained in the process of step S3. In the present embodiment, the parsing algorithm for machine translation is not a particular problem.

As shown in FIG. 10, the result of the syntax analysis performed by the syntax analysis unit 9 indicates that “a” is related to “student”, and “student” and “go to school” are related to “performed”. Is shown. Thus, the dependency relation of the input sentence can be understood. Step S5: The translation unit 11 converts the syntax analysis result obtained in the process of step S4 into a tree structure of English phrases by the translation dictionary 12 (the conversion result of FIG. 10), and creates an English sentence from this (FIG. 10). Translation result). This is the translation result. In this embodiment, the translation algorithm is not a particular problem.

Step S6: The machine translation result is displayed on the display unit 6.
The dictionary lookup result of the electronic dictionary 14 is displayed. Step S7: When performing the machine translation process and the dictionary lookup process of the next input sentence, the process returns to the step S1. If not, the process ends. Step S8: The electronic dictionary 14 is subjected to dictionary lookup using the character string of the word as a result of the morphological analysis of step S2 (the character string of which the utilization form is a terminal form) as a heading. This is illustrated in FIG.
This will be described in detail with reference to sub-flowcharts of FIG.

FIG. 3 is a dictionary lookup process (1) according to this embodiment.
6 is a flowchart showing the procedure of FIG. In FIG. 3, step S31: heading search unit 302 of heading acquisition unit 3
Thus, the search character string stored in the morphological analysis result buffer in the processing of step S2 is searched to obtain a heading (in the case of the inflected form, the heading of the ending form, and in the case of the inflected form, the heading as it is in the morphological analysis result).

FIG. 12 is a diagram showing an example of an input sentence and its dictionary headings according to this embodiment. As shown in FIG. 12, the result of the word division of the input sentence "Matter determined in the meeting" is "meeting", "de", "decided", "ta", "item", and the inflected character string. The final form of "decision" is "decision." (This is, when the character string “decision” is drawn by the morphological analysis dictionary 7 in step S2, this character string is a verb, a continuous use form utilizing five steps, and the final form is “decision”. I understand.)

Step S32: The search character string obtained in step S31 is stored in the index storage unit 301. Therefore, the input sentence "Matters decided in the meeting"
It is stored as each character string of “de”, “decide”, “ta”, and “item”. Step S33: All different notations acquisition section 30 of heading acquisition section 3
3, the dictionary search notation (heading) including all the different notations from the morphological analysis result obtained in the processing of step S2
Get. As shown in FIG. 11, “meeting”, “meeting”, and “meeting” are further obtained from “meeting”.

Step S34: The character string obtained in the process of step S33 is also stored in the heading storage unit 301. However, the same words as those stored in step S32 are omitted. In the example of FIG. 12, from the input sentence “Matter determined by meeting”
Character strings of “meeting”, “meeting”, “meeting”, “meeting”, “de”, “decide”, “ta”, and “item” are stored.

Step S35: In the dictionary search section 4,
Character strings stored in the heading storage unit 301 (in the case of the input sentence of FIG. 12, “meeting”, “meeting”, “meeting”,
The electronic dictionary 14 is searched using “meeting”, “de”, “decide”, “ta”, “item”) as a heading. The process proceeds to step S6 in the flowchart of FIG.

The display processing in step S6 may display all the searched character strings, or may display only the contents of a certain searched word by clicking a pointing device or the like. Regarding "meeting",
If "meeting" is in the heading of the electronic dictionary,
"Meeting", "in", "decide", "ta", "item"
The user is allowed to select one or more of them.

According to the above-described processing, just by inputting a sentence, the sentence is morphologically analyzed to extract a character string, and not only is the used word (variant form) corrected to a terminal form (original form, basic form), Even if it is different from the dictionary heading notation, all words included in the input sentence are used as electronic dictionary headings,
For example, when creating a foreign language, a user simply creates a sentence to be written in his / her own language, and searches a dictionary between a native language and another language (Japanese-English dictionary). Can concentrate on writing Japanese language sentences.

FIG. 4 is a dictionary lookup process (2) according to this embodiment.
6 is a flowchart showing the procedure of FIG. In FIG. 4, step S41 is the same as the processing of step S31. Step S42: The character string obtained in the process of step S41 is stored in the heading storage unit 301. However, a character string including a different notation is not stored (step S43, step S43).
It is left to 44 processing. That is, in the example of FIG. 12, the character strings stored in the heading storage unit 301 in the input sentence “item determined by the meeting” are “de”, “determined”, “ta”, and “item”.

Step S43: The different notation acquisition unit 304 of the entry acquisition unit 3 obtains a dictionary search notation (heading) that does not include the different notation from the morphological analysis result obtained in the process of step S2. In the example of FIG. 12, from FIG. 11, the notation for the dictionary search of the input character string “meeting” is “meeting”.
It is.

Step S44: The character string obtained in the process of step S43 is stored in the index storage unit 301. FIG.
In the example of 2, the result of step S42 and step S44,
The character strings stored are “meeting”, “de”, “decide”, “ta”, and “item”, as shown in the dictionary heading of FIG. Step S45: It is the same as the processing of step S35.

As shown in FIG. 11, it is possible to search the electronic dictionary by checking in advance what the title of the electronic dictionary is and storing it (indicated by ●). That is, the words described in FIG. 12 are dictionary headings. The morphological analysis unit that performs pre-processing of the dictionary lookup is also efficient because it is also used for machine translation, and it is possible to create a foreign language sentence by referring to both the result of the machine translation and the result of the dictionary lookup.

FIG. 5 is a dictionary lookup process (3) according to this embodiment.
6 is a flowchart showing the procedure of FIG. Step S51: It is the same as the processing of step S31. Step S52: It is the same as the processing of step S32. Step S53: The all-morpheme analysis result acquisition unit 305 obtains, in a basic form, a heading of the first place or less (a heading with a different position) from the character string stored in the morphological analysis result buffer in the processing of step S2. It is assumed that an algorithm (an algorithm widely used in machine translation) for searching for a plurality of headings is used in the morphological analysis processing in step S2.

FIG. 13 is a diagram showing an example of an input sentence and a result of morphological analysis thereof according to the present embodiment. In the case of FIG.
There is ambiguity in the processing of the character string part of "Kyotohon"
"Today", "book", "now", and "Japan" can be considered. At this time, if "now" and "Japan" are the first headline candidates, the second headline candidates "today" and "book" are obtained. Step S54: The character string obtained in the process of step S53 is stored in the index storage unit 301. In the example of FIG.
As a result of Steps S52 and S54, the stored character strings are “now”, “Japan”, “today”, “book”,
"To" and "to read."

Step S 55: The same as the processing in step S 35. Step S56: For the character string for which dictionary search was possible, the character string is displayed on the display unit 6. In the case of the input sentence "Read today's book" in FIG. 13, "now", "Japan", "today", "book",
"" And "read" are displayed. Step S57: The user selects a heading to be looked up in the dictionary by using a cursor key, a mouse or the like. The dictionary lookup result of the selected headline is displayed on the display unit 6 in step S6.

According to the dictionary lookup process (3), even if there is ambiguity in the word extraction of the input sentence, all the words included in the input sentence (however, notations not included in the headings of the electronic dictionary are excluded) ) In an electronic dictionary, a user who creates a foreign language only needs to create a sentence he wants to write in his own language.

FIG. 6 is a dictionary lookup process (4) according to this embodiment.
6 is a flowchart showing the procedure of FIG. In the example of FIG. 6, the heading of the morphological analysis dictionary 7 and the heading of the electronic dictionary 14 are checked in advance, the heading of the morphological analysis dictionary 7 is not in the electronic dictionary 14 as it is, When a part or the whole is composed of one or more headings of the electronic dictionary 14, the compound heading correspondence table 13 in which the headings of the morphological analysis dictionary 7 and the one or more headings of the electronic dictionary 14 are stored as a pair is stored. Have it ready.

FIG. 14 is a diagram showing an example of a compound heading stored in the compound heading correspondence table according to this embodiment.
In the example of FIG. 14, the compound heading “syntax analysis” is treated as one word in machine translation, and is often translated as “parser”. (Dictionaries for machine translation are relatively frequently registered in long units in order to improve accuracy.) However, in general Japanese-English dictionaries, one word cannot be closed, and "syntax" and "syntax"analysis"
("Analysis"). This means that the result of the morphological analysis in step S2 is “syntax analysis”, but the electronic dictionary 14 is referred to by “syntax” and “analysis”.

Step S61: The same as step S31. Step S62: The same as step S32. Step S63: The compound heading correspondence table search unit 306 of the heading acquisition unit 3 searches the compound heading correspondence table 13 with the character string obtained in step S61.

Step S64: The headline retrieved in step S63 is the combined headline acquisition unit 307 of the headline acquisition unit 3.
To obtain a corresponding heading from the composite heading correspondence table 13. For example, if the result of the morphological analysis in step S2 is “syntax analysis”, the headings of the electronic dictionary 14 are “syntax” and “analysis” in the composite heading correspondence table of FIG.
It is.

Step S65: The character string obtained in step S64 is stored in the index storage unit 301, and the original compound word is deleted. For example, if the result of the morphological analysis in step S2 is “syntax analysis”, the “syntax analysis” is stored in the index string storage unit 301 in step S62, but the electronic dictionary 14 cannot be searched in the “syntax analysis”. Therefore, it is deleted from the header string storage unit 301, and “syntax” and “analysis” are used instead.
Is stored. Step S66: Same as step S35.

According to the dictionary lookup processing (4), even if the extracted word is a compound word not found in the electronic dictionary as a result of the morphological analysis, it is analyzed into words constituting the compound word and the dictionary is looked up. A user who creates a foreign language only needs to create a sentence to be written in his or her own language.

FIG. 7 shows dictionary lookup processing (5) according to this embodiment.
6 is a flowchart showing the procedure of FIG. Step S71: It is the same as the processing of step S31. Step S72: It is the same as the processing of step S32.

Step S73: The character strings stored in the search character string storage unit 301 in the processing of step S72 are sequentially set as primary keys. When all strings are primary keys, it ends,
Otherwise, use the next string as the primary key. Step S74: Similar to step S35, the dictionary search unit 4
, The electronic dictionary 1 is searched using the primary key. Step S75: The secondary key table search section 501 searches the secondary key table 15 with the primary key.

FIG. 15 is a diagram showing an example of the primary key and the secondary key stored in the secondary key table according to the present embodiment. FIG.
In the case of 5, when the primary key is “head”, the alternative key candidates are “hard”, “come”, “use”, and “low”. Step S76: Character string acquisition unit 502 from input character string
In step S72, a character string other than the character string currently used as the primary key is obtained from the heading stored in the heading storage unit 301 in step S72. Alternatively, a character string whose distance from the primary key in the input character string exists within n words is obtained.

In the input sentence “There is a man with a hard head”, if the primary key is “head” and the condition of the primary key and the sub key is “up to one word”, the obtained character string is Hard. "
They are the particle "no" and "male" with one word "no" between them. Step S77: The sub key table comparison unit 503 compares the sub key candidate obtained from the sub key table 15 in step S75 with the character string obtained in step S76. If there is a matching character string, use it as a secondary key. In the example, "hard" is the secondary key.

Step S78: In the description contents of the electronic dictionary 14 searched by the primary key, the position of the secondary key is searched from the secondary heading or the example. The search method may be a simple search method by comparing the character strings by shifting the subheadings or examples of the description contents of the electronic dictionary 14 searched by the primary key one character at a time. The secondary key table may be extended to store the appearance location (the nth line, etc.) of the secondary key in the dictionary contents of the primary key. Step S79: The secondary key display processing unit 504 searches the primary key of the dictionary for a subheading that includes a character string that matches the secondary key or an item that has an example that includes a character string that matches the secondary key. Processing to display at the top of the screen or at the center of the screen is performed.

FIG. 16 is a diagram showing an example of a screen of the dictionary lookup process (5) according to this embodiment. In the example of FIG. 16, when the primary key is “head” and the sub key is “hard”, the subheading “hard” is displayed at the top of the screen (window). In addition to being displayed at the center of the screen or at the top of the screen, other sub-headings are displayed only in the heading part, and other meaning items are displayed only in the parsing part. Of course, instead of being displayed at the center of the screen or at the top of the screen, other sub-headings may be displayed only in the heading part, and other meaning items may be displayed only in the transliteration part (see FIG. 8).

In this step, only the display image is created, and the display itself is performed by the display unit 6 in step S6. It returns to step S73.

According to the dictionary lookup processing (5), not only the words that appear in the input sentence are looked up in the dictionary, but also the words before and after the words that are most useful in the description contents of the headings of the electronic dictionary. Is displayed first, so that you can find the reference point faster.

FIG. 8 is a dictionary lookup process (6) according to this embodiment.
6 is a flowchart showing the procedure of FIG. This corresponds to the step S79. Step S81: The same processing as the contents shown in step S79 is performed. In this embodiment, the subheading is displayed at the center of the screen or at the top of the screen, the other subheadings are displayed only at the heading part, and other meaning items are displayed only at the transliteration part. However, without displaying at the center of the screen or at the top of the screen,
Other subheadings are only the heading part, other semantic items are
Only the transliteration part may be displayed.

FIG. 18 is a diagram showing an example of a screen of the dictionary lookup process (6) according to this embodiment. The example of FIG. 18 is an example of a primary key “head” and a secondary key “low”. The place where the first subkey appears is an example of the meaning [I]. Therefore,
The meaning [I] displays the entire description, the other meanings [I] and [II] display only the transliteration part (in parentheses), and the idiom of the subheading displays only the heading part.

Step S6: The contents prepared in step S81 are displayed on the display unit 6. Step S82: It is checked which of the special cursor keys 16 has been pressed. If it is the next key, go to step S83. If it is the previous key, go to step S84. If it is the first key, go to step S8.
Go to 5.

Step S83: The next subkey appearance location is set at the center of the screen or at the top of the screen. If there are no more occurrences of the alternate key, a warning is issued. In this example,
The same processing as in step S81 is performed. In FIG. 18, the second appearance location is “head is low” which is a subheading. Therefore, all the contents of the subheading "low head" are displayed, and the other subheadings "hard head", "cut head", "cool head" and meanings [I] and [II] Only the outline is displayed (the meaning item is the transliteration part, and the subheading is the heading part) (this state is not shown in FIG. 18).

Step S84: The previous sub key appearance location is set to the center of the screen or the top of the screen. If there are no more secondary key occurrences, a warning is issued. In this embodiment, the same processing as in step S81 is performed. Step S85: The head of the dictionary description contents of the primary key is set.
All may be displayed in summary or all may be displayed in full.

According to the dictionary lookup processing (6), not only the words appearing in the input sentence are merely looked up in the dictionary, but the words before and after the words are most useful in the description contents of the corresponding headline in the electronic dictionary. Since the part is displayed first or the part other than the relevant part is displayed as a summary, the reference part can be found more quickly.

FIG. 9 is a dictionary lookup process (7) according to this embodiment.
6 is a sub-flowchart showing the procedure of FIG. Step S79 in FIG. 7 is replaced with step S91. Step S91: Of the dictionary contents of the primary key, a character string that matches the secondary key is characteristically displayed by color, underline, box, etc. as shown in FIG.

According to the dictionary lookup process (7), not only the words appearing in the input sentence are simply looked up in the dictionary, but the words before and after the words are used to display the characteristic display contents of the headline in the electronic dictionary. , So you can find the reference point faster.

[0077]

According to the present invention, even if a character string of an input sentence is different from a heading of an electronic dictionary due to a difference in a sentence kana, a notation or the like, a character string obtained by morphologically analyzing the sentence is used as a heading and is associated with the character string. Since all of the different notations are acquired and the conjugate words are corrected to the basic form, it is possible to search the electronic dictionary with a high probability only by inputting sentences. For example, when translating a sentence in one's own language into a sentence in another language, a Japanese-English dictionary can be searched simply by inputting a sentence written in one's own language. You can concentrate on translating into other languages.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a basic configuration of an electronic dictionary search device of the present invention.

FIG. 2 is a main flowchart showing a procedure of a translation process / dictionary lookup process according to the embodiment.

FIG. 3 is a sub-flowchart showing a procedure of a dictionary lookup process (1) according to the embodiment.

FIG. 4 is a sub-flowchart showing a procedure of a dictionary lookup process (2) according to the embodiment.

FIG. 5 is a sub-flowchart showing a procedure of a dictionary lookup process (3) according to the embodiment.

FIG. 6 is a sub-flowchart showing a procedure of a dictionary lookup process (4) according to the embodiment.

FIG. 7 is a sub-flowchart showing a procedure of a dictionary lookup process (5) according to the embodiment.

FIG. 8 is a sub-flowchart showing a procedure of a dictionary lookup process (6) according to the embodiment.

FIG. 9 is a sub-flowchart showing a procedure of a dictionary lookup process (7) according to the embodiment.

FIG. 10 is an explanatory diagram showing a procedure of a Japanese-English machine translation process according to the embodiment.

FIG. 11 is a diagram showing an example of deformed display information and a dictionary reference mark stored in the morphological analysis dictionary according to the embodiment.

FIG. 12 is a diagram showing an example of an input sentence and its dictionary heading according to the embodiment.

FIG. 13 is a diagram illustrating an example of an input sentence and a morphological analysis result thereof according to the present embodiment.

FIG. 14 is a diagram illustrating an example of a compound heading stored in a compound heading correspondence table according to the embodiment;

FIG. 15 is a diagram illustrating an example of a primary key and a secondary key stored in a secondary key table according to the embodiment.

FIG. 16 is a diagram illustrating a screen example of a dictionary lookup process (5) according to the embodiment.

FIG. 17 is a diagram illustrating a screen example of a dictionary lookup process (7) according to the embodiment.

FIG. 18 is a diagram illustrating a screen example of a dictionary lookup process (6) according to the embodiment.

FIG. 19 is a block diagram illustrating one configuration of a heading acquisition unit according to the embodiment.

FIG. 20 is a block diagram illustrating one configuration of a display processing unit according to the present embodiment.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Input part 2 Morphological analysis part 3 Heading acquisition part 4 Dictionary search part 5 Display processing part 6 Display part 7 Morphological analysis dictionary 8 Analysis result buffer 9 Syntax analysis part 10 Syntax analysis dictionary 11 Translation part 12 Translation dictionary 13 Compound heading Correspondence table 14 Electronic dictionary 15 Secondary key table 16 Special cursor key 301 Heading storage unit 302 Heading search unit 303 All different notation acquisition unit 304 Different notation acquisition unit 305 All morphological analysis result acquisition unit 306 Compound heading correspondence table search unit 307 Compound heading acquisition Unit 501 secondary key table search unit 502 input character string acquisition unit 503 secondary key table comparison unit 504 secondary key display processing unit 505 feature display processing unit 506 summary display processing unit

Claims (9)

[Claims] An electronic dictionary storing headings and candidates corresponding thereto, an input unit for inputting a sentence and instructing a dictionary search, and a morphological analysis based on a morphological analysis dictionary for the sentence input by the input unit A morphological analysis unit, a heading obtaining unit that obtains a character string morphologically analyzed by the morphological analysis unit as a heading, corrects the inflected words of the obtained heading to a basic form, and stores it in a heading storage unit; and A dictionary search unit that searches and converts a heading stored in the heading storage unit to a corresponding candidate, and a display unit that displays an input sentence and the dictionary searched candidate, the character string obtaining unit includes: An electronic dictionary search device characterized by acquiring all of different notations related to a character string morphologically analyzed by a morphological analysis unit as headings. 2. The method according to claim 1, wherein the heading acquisition unit is configured to search the electronic dictionary for a notation that matches a heading stored in the electronic dictionary among different notations related to the character string morphologically analyzed by the morphological analysis unit. 2. The electronic dictionary search device according to claim 1, wherein a dictionary searchable mark indicating that there is is provided and stored in the index storage unit. 3. When the character string morphologically analyzed by the morphological analysis unit has a plurality of delimiter positions, the heading acquisition unit converts all of the character strings delimited by the plurality of delimiter positions into headings of the electronic dictionary. The electronic dictionary search device according to claim 1, wherein the electronic dictionary search device is obtained as a search result and stored in the index storage unit. 4. A compound heading correspondence table in which compound words each having a set of a plurality of headings stored in the electronic dictionary are stored in advance, and the heading acquisition unit searches the compound heading correspondence table to retrieve the morpheme. If the character string subjected to morphological analysis by the analysis unit includes the compound word, a heading constituting the compound word is acquired as a heading of the electronic dictionary and stored in the heading storage unit. An electronic dictionary search device according to claim 1. 5. A sub-key table in which a main key and a sub-key are stored in advance using a heading of the electronic dictionary as a primary key and an example phrase thereof as a sub-key, and a character string searched dictionary by the dictionary search analysis unit. When there is a key heading, a sub-key table search unit for searching the sub-key table for a sub-key using a primary key, and an example word that matches the sub-key searched by the sub-key table search unit are displayed on a predetermined screen. 2. The electronic dictionary search device according to claim 1, further comprising: a display processing unit for displaying a location. 6. A sub-key designating unit for designating a sub-key searched by the sub-key table search unit, wherein the display processing unit displays the sub-key designated by the sub-key designating unit at a predetermined position on a screen. 6. The electronic dictionary search device according to claim 5, wherein the electronic dictionary is displayed. 7. The electronic dictionary search device according to claim 5, wherein the display processing unit displays a primary key corresponding to a heading of the electronic dictionary and a sub key associated with the primary key so as to be identifiable. . 8. The electronic dictionary search device according to claim 5, wherein the display processing unit displays a list of sub headings of the electronic dictionary that match the sub key. 9. A program for controlling, by a computer, an electronic dictionary search device including an electronic dictionary storing headlines and candidates corresponding thereto, an input unit for inputting text and instructing dictionary search, and a display unit. Media
The computer makes the sentence input by the input unit morphologically analyzed based on the morphological analysis dictionary, acquires the morphologically analyzed character string and all the different notations related to the character string as headings, and obtains the obtained heading. The corrected word is corrected to a basic form and stored in the heading storage unit, the electronic dictionary is searched, and the heading stored in the heading storage unit is converted into a corresponding candidate. A medium storing an electronic dictionary search device control program to be displayed on the display unit.