JP2005135444A - Character string conversion device having proofreading support function and its method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a character string conversion device capable of converting a character string while supporting proofreading. <P>SOLUTION: A kana character string provided for a conversion means 2 is stored in a kana character string storage part 14. A segment creation means 14 creates a feasible segment from the kana character string stored in the storage means 14 by referring to a dictionary 12, and store it in a segment candidate storage part 6. A segment candidate selection means 16 finds out a feasible segment path by combining the segments stored in the storage part 6, and selects one of them as a converted character string candidate. A proofreading support means 8 creates proofreading support information about the converted character string candidate based on the part of speech of the segment stored in the storage part 6 and outputs it to a display means 20 to display it. The proofreading support means also creates proofreading support information based on proofreading-related information relating to the segment stored in the storage part 6 and outputs it to the display means 20 to display it. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to character string conversion, and more particularly to a technique for supporting proofreading of the converted character string.

A system for converting a given kana character string into a character string including kanji has been put into practical use. This system is used in a Japanese word processor or the like as a system (kana-kanji conversion system) for inputting kana characters and converting them into kanji-kanji sentences.

In such a kana-kanji conversion system, it is determined by the user to use whether or not the converted kanji magic sentence is appropriate as Japanese. That is, it is determined whether or not the text is appropriate while the text is input using the Kana-Kanji conversion system, or whether the text is appropriate after the text is printed.

However, the calibration work as described above is not always easy, and there are cases where sufficient calibration cannot be performed depending on the user.

SUMMARY OF THE INVENTION An object of the present invention is to provide a character string conversion apparatus and method that can solve the conventional problems as described above and can easily perform proofreading.

Means to solve the problem

(1) A character string conversion device according to the present invention includes a dictionary that stores a conversion character string and its part of speech in association with a kana character string, and refers to the dictionary to divide a given kana character string into phrases. Output conversion character string candidates converted into conversion character strings, receive conversion commands and output selected conversion character string candidates as fixed character strings, and the part of speech of each phrase in the conversion character string candidates And a calibration support means for performing an output for calibration support.

(2) In the character string conversion device according to the present invention, the proofreading support means determines whether or not an adjunct in each phrase in the converted character string candidate is a particle, and the phrase having the same particle continues for a predetermined number of times. In such a case, the proofreading support output indicating that the particles are continuous is performed.

(3) In the character string conversion device according to the present invention, the proofreading support means determines whether or not each phrase is in a combination modification usage in the conversion character string candidate, and the clause of the combination modification usage is a predetermined number of times. When it continues, the proofreading support output that the modifier is continuous is performed.

(4) In the character string conversion device according to the present invention, the proofreading support means includes a phrase constituted by any one of a) a conjunction, b) a conjunction form of a predicate, c) a noun equivalent and a particle equivalent. When found, this clause determines that the application method of the combination modification has been continued once, and after this clause, a) a conjunction, b) a combination of predicates, c) a combination of a noun equivalent and a particle equivalent, d) It is characterized in that the continuation of the phrase of the combination modification usage is determined based on how many times the phrase composed of any combination of the noun and the particle “no” continues.

(5) In the character string conversion device according to the present invention, the proofreading support means determines whether the sentence is normal or respected based on the part of speech of the phrase in the converted character string candidate, and based on the determination It is characterized in that it performs calibration support output.

(6) In the character string conversion device according to the present invention, the proofreading support means, when the conversion character string candidate clause is “is”, “mas”, “has”, or other utilization forms of these, It is characterized by judging that it is a respectful body when it includes “please”, “please”, and “please”.

(7) In the character string conversion device according to the present invention, the proofreading support means is normal when the phrase immediately before the punctuation mark or symbol among the phrases of the converted character string candidate is one of the following: It is characterized by judging.

a) Combination of the body and final particle "ka" b) Combination of body and "no" and final particle c) When the last phrase or auxiliary verb is imperative d) The last sentence of the phrase is connected to the final form In the case of a combination of particles, e) In the case where the last sentence of the clause is a combination of a final form and a final particle, f) In the case where the last sentence of the phrase is a combination of a combined form and a connected particle, g) In the case of a combination of final form and final particle h) In the case where the last sentence of the phrase is a combination of the combined form, “no” and final particle
(8) In the character string conversion device according to the present invention, when the predetermined “normal” or “respectful” style and the style determined by the proofreading support means are different, the style is different. It is characterized by output.

(9) In the character string conversion device according to the present invention, when the proofreading support means determines that the sentence style of each phrase is not the same with respect to one sentence converted character string candidate, the proofreading that the style has changed It is characterized by performing support output.

(10) The character string conversion device according to the present invention is characterized in that the proofreading support means outputs a proofreading support output indicating that it is a passive expression when the phrase includes auxiliary verbs “de” and “being”.

(11) A character string conversion device according to the present invention includes a dictionary that stores a conversion character string and proofreading-related information in association with a kana character string, and refers to the dictionary to convert a given kana character string into a phrase. A conversion means that outputs a conversion character string candidate that is separated and converted into a conversion character string, and that receives a confirmation command, and outputs the selected conversion character string candidate as a confirmation character string, and the conversion means correspond to each phrase And proofreading support means for performing output for proofreading support based on the proofreading related information obtained from the dictionary.

(12) The character string conversion device according to the present invention displays a conversion character string candidate for a given kana character string on the display screen, and also displays or voices the calibration support information for the conversion character string candidate. It is characterized by that.

(13) The character string conversion device according to the present invention is characterized in that the proofreading support information is displayed in a line near the conversion character string candidate and different from the conversion character string candidate.

(14) The character string conversion device according to the present invention is characterized in that the proofreading support information is displayed on the same line as the character string candidate.

(15) The character string conversion device according to the present invention is characterized in that the proofreading support information is displayed at a predetermined position irrespective of a predetermined display position of the character string candidate.

(16) The character string conversion device according to the present invention displays a conversion character string candidate for a given kana character string on the display screen, and displays that there is proofreading support information for each clause of the conversion character string candidate. In addition, the proofreading support information is displayed only when the cursor is moved to a phrase for which proofreading support information is displayed.

(17) In the character string conversion method according to the present invention, the conversion character string and its part of speech are stored in a dictionary in advance in association with the kana character string, and the given kana character string is referred to the dictionary, In the character string conversion method for outputting a conversion character string candidate converted into a conversion character string by dividing into clauses, and receiving a confirmation command, the selected conversion character string candidate is output as a confirmation character string. It is characterized in that output for proofreading support is performed based on the part of speech of each clause.

(18) In the character string conversion method according to the present invention, it is determined whether or not an adjunct in each phrase in the converted character string candidate is a particle, and if a phrase having the same particle continues for a predetermined number of times, the particle is It is characterized in that it provides a calibration support output indicating that it is continuous.

(19) In the character string conversion method according to the present invention, in the converted character string candidate, it is determined whether or not each clause is a combination modification usage, and when the clause of the combination modification usage continues for a predetermined number of times, A proofreading support output indicating that modifiers are continuous is performed.

(20) In the character string conversion method according to the present invention, based on the part of speech of the phrase in the converted character string candidate, it is determined whether the sentence is normal or respected, and the proofreading support output is output based on the determination. It is characterized by what it does.

(21) The character string conversion method according to the present invention is characterized in that, when the phrase in the conversion character string candidate includes auxiliary verbs “de” and “being”, proofreading support output indicating passive expression is performed.

(22) In the character string conversion method according to the present invention, the conversion character string and the proofreading related information are stored in advance in a dictionary in association with the kana character string, and the kana character string is stored in a phrase by referring to the dictionary. In the character string conversion method that outputs a conversion character string candidate that is delimited and converted into a conversion character string, and receives a confirmation command, and outputs the selected conversion character string candidate as a confirmation character string, the dictionary is associated with each clause. Based on the calibration-related information obtained, output for calibration support is performed.

(23) In the character string conversion method according to the present invention, the conversion character string candidate for the given kana character string is displayed on the display screen, and the proofreading support information for the conversion character string candidate is also displayed or output in voice. It is characterized by that.

The terminology used to describe this invention is as follows.

The “kana character string” is a concept including not only a character string including only a kana character string but also a character string having alphabets, katakana, symbols, and the like.

The “conversion character string” refers to a character string composed of kanji, kana, katakana, alphabet, symbols, etc. alone or in combination.

“Conversion character string candidate” refers to a candidate for a conversion character string. It is a concept that includes not only all indeterminate states but also some that have definite portions.

“One sentence” refers to one sentence that ends with a punctuation mark or the like.

The “cursor” refers to a display on the screen for clarifying the input area and the control target area. The attention phrase cursor CK in the embodiment corresponds to this.

“Output of proofreading support information” means displaying or printing the proofreading support information or giving it to other software as data. Further, it is a concept including notifying a human by voice or vibration.

The “storage medium” refers to a computer-readable medium such as a flexible disk, a hard disk, or a CD-ROM that can fix a program.

“Computer-executable program” can be executed not only when the program stored in the storage medium can be directly executed, but also when combined with a program that can be executed by installation or other programs. Indirectly executable programs such as programs are also included.

(1) (17) The character string conversion device and the character string conversion method according to the present invention are characterized in that output for proofreading support is performed based on the part of speech of each phrase in the converted character string candidate. . Accordingly, it is possible to provide proofreading support information during the character string conversion operation.

(2) (18) The character string conversion device and the character string conversion method according to the present invention perform proofreading support output indicating that a particle is continuous when a phrase having the same particle continues for a predetermined number of times. It is characterized by that. Therefore, an inappropriate continuous state of particles can be given as proofreading support information at the time of character string conversion operation.

(3) (19) The character string conversion device and the character string conversion method according to the present invention indicate that, in the conversion character string candidate, if the clauses of the conjunctive modification usage continue for a predetermined number of times, the modifier is continuous. It is characterized by performing calibration support output. Therefore, an inappropriate continuous state of modifiers can be given as proofreading support information during a character string conversion operation.

(4) In the character string conversion device according to the present invention, the proofreading support means includes a phrase constituted by any one of a) a conjunction, b) a conjunction form of a predicate, c) a noun equivalent and a particle equivalent. When it is found, it is determined that the application method of the combination modification is continued once by this clause, and after the clause, a) a conjunction, b) a combination of a predicate, c) a combination of a noun equivalent and a particle equivalent, d) It is characterized in that the continuation of the clauses of the usage of the combination modification is determined based on how many times the clause composed of the combination of the noun and the particle “no” continues. Therefore, it is possible to detect a series of inappropriate modifiers more accurately.

(5) (20) The character string conversion device and the character string conversion method according to the present invention determine whether a sentence is normal or respected based on the part of speech of the phrase in the conversion character string candidate, It is characterized in that the calibration support output is performed based on the judgment. Therefore, it is possible to give the style as proofreading support information during the character string conversion operation.

(6) In the character string conversion device according to the present invention, the phrase of the conversion character string candidate is “is”, “mass”, “is”, or other utilization forms of these, or “please” “please” It is characterized by judging that it is a respectful body when it includes “Otakusei”. Therefore, it is possible to detect the honorable body more accurately.

(7) The character string conversion device according to the present invention is characterized in that, among the phrases of the converted character string candidates, the sentence immediately before the punctuation mark or symbol is determined to be normal when the phrase is any of the following: It is said. a) a combination of the body and the final particle “ka”, b) a combination of the body and the “no” and the final particle, c) if the last sentence or auxiliary verb is imperative, and d) the last sentence of the section is terminated. In the case of a combination of a form and a connective particle, e) In the case where the last sentence of the clause is a combination of a final form and a final particle, f) In the case where the last sentence of the clause is a combination of a combined form and a particle, g) When the last word is a combination of a final form and a final particle, h) When the last sentence of a phrase is a combination of a combined form, “no”, and a final particle. Therefore, the normal substance can be detected more accurately.

(8) In the character string conversion device according to the present invention, when the predetermined “normal” or “respectful” style and the style determined by the proofreading support means are different, the style is different. It is characterized by output. Therefore, it is possible to give, as proofreading support information, that a sentence style different from the scheduled style is input during the character string conversion process.

(9) The character string conversion apparatus according to the present invention performs a calibration support output indicating that the style of the sentence has changed when it is determined that the style of each phrase is not the same for a converted text string candidate of one sentence. It is characterized by. Accordingly, it is possible to provide the proofreading support information that a sentence with a changed style is input during the character string conversion process.

(10) (21) In the character string conversion device and the character string conversion method according to the present invention, when the phrase in the conversion character string candidate includes the auxiliary verbs “being” and “being”, the proofreading support output that indicates passive expression is performed. It is characterized by that. Therefore, at the time of character string conversion processing, information indicating passive expression can be given as proofreading support information.

(11) (22) The character string conversion device and the character string conversion method according to the present invention perform output for proofreading support based on proofreading related information obtained from a dictionary in association with each phrase. It is characterized by. Therefore, the proofreading support information can be given based on the information obtained from the dictionary during the character string conversion process.

(12) (23) The character string conversion device and the character string conversion method according to the present invention are characterized in that the proofreading support information for the converted character string candidates is displayed or output by voice. Therefore, it is possible to easily obtain proofreading support information corresponding to the display of the conversion character string candidate.

(13) The character string conversion device according to the present invention is characterized in that the proofreading support information is displayed in a line near the conversion character string candidate and different from the conversion character string candidate. Therefore, it is easy to visually confirm the calibration support information.

(14) The character string conversion device according to the present invention is characterized in that the proofreading support information is displayed on the same line as the character string candidate. Therefore, it is easy to visually confirm the calibration support information.

(15) The character string conversion device according to the present invention is characterized in that the proofreading support information is displayed at a predetermined position irrespective of a predetermined display position of the character string candidate. Therefore, the display position of the calibration support information is fixed and easy to confirm.

(16) The character string conversion device according to the present invention displays that the proofreading support information exists for each phrase of the conversion character string candidate, and the cursor is moved to the phrase for which proofreading support information is displayed. This is characterized in that the proofreading support information is displayed only in the case of the case. Therefore, the presence / absence of the calibration support information can be easily confirmed, and the display necessary range of the contents can be reduced.

BEST MODE FOR CARRYING OUT THE INVENTION

FIG. 1 shows the overall configuration of a character string conversion apparatus according to an embodiment of the present invention. In this embodiment, the kana character string storage unit 4, the phrase candidate storage unit 6, the phrase generation unit 14, and the phrase candidate selection unit 16 constitute the conversion unit 2. Further, a dictionary 12 is constituted by the independent word dictionary 8 and the attached word dictionary 10.

The “kana character string” given to the conversion means 2 is stored in the kana character string storage unit 14. The phrase generation unit 14 refers to the dictionary 12 and generates a possible phrase from the “kana character string” stored in the kana character string storage unit 14. The phrase generation unit 14 stores the generated phrase in the phrase candidate storage unit 6. That is, the kanji, the part of speech, the utilization form, the proofreading related information, etc. acquired from the dictionary for the independent words and the attached words of each phrase are stored. The phrase candidate selection means 16 finds possible phrase paths by combining the phrases stored in the phrase candidate storage unit 6. When there are a plurality of possible phrase paths, the phrase path (converted character string candidate) is narrowed down to one by a predetermined selection process.

The phrase candidate selection means 16 sends this converted character string candidate to the display means 20 for display. Further, when a change command is given, different conversion character string candidates are displayed on the display means 20. When a confirmation command is given, the converted character string candidate is output as a confirmed character string.

The proofreading support means 18 generates proofreading support information based on the part of speech of the phrase stored in the phrase candidate storage unit 6 and outputs it to the display means 20 for display. Further, based on the proofreading related information related to the phrase stored in the phrase candidate storage unit 6, the proofreading support information is generated and output to the display means 20 for display.

FIG. 2 shows a hardware configuration when the character string conversion apparatus of FIG. 1 is realized using a CPU. Connected to the bus line 34 are a CPU 22, a hard disk 24, a display 20 as a display means, a memory 28, a flexible disk controller (FDD) 30, and a keyboard 32. The hard disk 24 stores an independent word dictionary 8, a character string conversion processing program 40, a proofreading support processing program 42, and the like. These are installed from the flexible disk (FD) 36 via the FDD 30. Of course, it may be taken from a CD-ROM or the like. Alternatively, it may be downloaded via a communication line.

The memory 28 is provided with a kana character string storage unit 4 for storing kana character strings and a phrase candidate storage unit 6 for storing phrase candidates. An attached word dictionary 10 is also provided.

FIG. 3 shows the data structure of the independent word dictionary 8. An independent word is a word that can be a phrase by itself. The independent word dictionary 8 stores kanji corresponding to the character string (reading) to be converted. Corresponding to these, parts of speech and proofreading related information are stored. Note that kana characters are generally stored in the “reading” column, but some or all of them may include alphabets. In the “kanji” column, not only kanji but also kana and alphabets may be stored.

FIG. 4 shows the data structure of the attached word dictionary 10. An adjunct is a word that cannot be a phrase by itself. The ancillary dictionary 10 stores an ancillary word, its corresponding part of speech, and a utilization form correspondingly. The parts of speech are classified and stored as particles, auxiliary verbs, and idioms. The particles are further classified and stored as case particles, connection particles, adjunct particles, and final particles. In addition, with respect to the endings of the idioms, the utilizable forms (pre-form, continuous form, end form ... instruction form) are stored. Furthermore, it is also remembered which word each appendage can be used after (an independent word / adjunct that can precede).

The part of speech classification used in this embodiment is shown below.

General noun: A general noun noun that can be attached to an object or the state of an object. Noun noun that can be suffixed with the variable "suru". Movement: A noun independent word that expresses the form, state, nature, etc. of each object and can be followed by “da” or “na”: a verbal conjunction that does not have a verb or auxiliary verb and has no connection with other words: Word conjunctives that are not used and that only modify body words: words that are used to continue phrases Phrases that express impressions, responses, calls, etc. Prefixes that are used before other words Suffixes that are used before other words: after other words Words used in conjunction with numbers: Words that express quantity, order, etc. Verbs: Further subcategorized according to how they are used Adjectives: Words that represent the nature and state of the word Adjective verbs: Properties and status of things The stands, word form when assert ends with "it,""Donot" adverb: word primarily to modify the verb - outline of the conversion processing and calibration support processing -
FIG. 5 shows a flowchart of the character string conversion program 40 and the proofreading support program 42. Both programs may be integrated, or a part or all of them may be modularized and separated. According to this flowchart, the process of converting the kana character string "I have Japanese language" will be explained. First, when the kana character string is input from the keyboard 32 (step S1), the CPU 22 stores it in the kana character string storage unit 4 in the memory 28 (see FIG. 6).

Next, the CPU 22 searches the independent word dictionary 8 and the attached word dictionary 10 in order from the first kana character in the kana character string in the kana character string storage unit 4 to generate a phrase (step S2). That is, all possible phrases are generated and stored in the phrase candidate storage unit 6 in the memory 28. FIG. 6 shows the phrases stored in the phrase candidate storage unit 6 in this way in association with the kana character string. In the figure, α1, α2, α3, α4, β1,... Are generated. The phrases α1 to α4 are kanji (and attached words) corresponding to the search character string starting with “wa”. The phrase α1 has a search character string of one character, the phrase α2 has two characters, the phrase α3 has three characters, and the phrase α4 has four characters. For each length of search character string, only one phrase is stored as a candidate. In other words, in the phrase α1, many kanji characters are searched in addition to “sum”, but only one kanji “sum” having the highest priority is selected in consideration of the frequency of use and the relationship with the preceding and following words. doing. The same applies to the phrases α2, α3, and α4.

The phrases β1, β2, and β3 are kanji (and attached words) corresponding to the search character string starting with “ta”. Similarly, all possible phrases are stored in the phrase candidate storage unit 6. When storing, as shown in FIG. 7, the part of speech, utilization form, and proofreading related information (described later) acquired from the dictionary are also stored. The phrase position information, that is, the information corresponding to which part of the kana character string is stored, but is omitted in FIG.

When the phrase generation and the phrase candidate storage unit 6 are completed as described above, the CPU 22 selects a phrase route candidate (step S3). In other words, it finds clauses that can be combined in association with kana character strings. For example, clauses β1 to β3 are the only clauses that can be continued with the “sum” of clause α1, and other clauses cannot be continued. This kind of consideration is performed up to the last phrase, and possible combinations of phrases (called phrase paths) are found. When a plurality of phrase paths are found, one of the highest priority phrase paths is selected based on the learning information and the relation between phrases. By selecting the phrase path in this manner, for example, a conversion character string candidate “I am / Japanese / processing / impressed / is /.” Is selected (/ represents a section break).

Next, the CPU 22 determines the necessity of proofreading support based on the part of speech of each phrase of the converted character string candidate (step S4). If proofreading support is necessary, proofreading support information is also displayed when the converted character string candidate is displayed on the display 20 (step S5). If there is no need for calibration support, only the converted character string candidates are displayed on the display 20 (step S6). A display example is shown in FIG. 8A. In the figure, the phrase “I am” is surrounded by a cursor CK. This indicates that the phrase that is the current processing target (referred to as the focused phrase) is “I am”. Note that the process of determining whether or not calibration support is necessary will be described in detail later.

After displaying the converted character string candidates, the process waits for a command from the user (step S7). When the user sees the converted character string candidate displayed on the display 20 and inputs various commands from the keyboard 32 (which may be a mouse or the like), in step S7, when the command for moving the noticeable phrase is input, the CPU 22 Moves the target phrase according to the commanded direction, and also moves the display of the cursor CK. For example, in the state of FIG. 8A, when a command to move the target phrase to the right is given, the cursor CK moves to the right and “Japanese” becomes the target phrase as shown in FIG. 8B. Further, when an instruction to move to the right twice is given, as shown in FIG. 8C, the cursor CK of the target phrase moves to “I'm impressed”.

In step S7, when an instruction for another candidate is given, other candidates are selected from the independent word dictionary 8 and the attached word dictionary 10 for the independent word and the attached word of the noticeable phrase “Kansei ga” (step S10). For example, select “I am interested”. At this time, the CPU 22 rewrites the “impressive” part of the phrase candidate storage means 6 shown in FIG. 7 into “interested”, and newly acquires part-of-speech, usage forms, proofreading related information, and the like from the dictionary. Rewrite to Thereafter, the new conversion character string candidate is again checked for the need for proofreading support (step S4) and displayed (steps S5 and S6). That is, a display as shown in FIG. 8D is performed.

When a post-conversion command is given in step S7, the entire noticeable phrase is post-converted according to the content of the command (step S12). For example, when a post-conversion command to katakana is given, the entire noticeable phrase is changed to “kansinga” as shown in FIG. 8E. In response to this, the contents of the phrase candidate storage unit 6 in FIG. 7 are also rewritten in the same manner as described above. However, since this word is not obtained from the dictionary (referred to as post-conversion word), it is forced to be a general noun in the part of speech column.

When a phrase segment change command is given in step S7, the target phrase is lengthened (or shortened) (step S11), and a phrase path candidate is selected again based on the stored contents of the phrase candidate storage unit 6. (Step S3). That is, a new converted character string candidate having a different phrase break position is generated. The subsequent processing is the same as the processing described above, and after determining whether or not the calibration support is necessary again, display is performed (steps S4, S5, and S6).

When the confirmation command is input through the above processing, the converted character string candidate at that time is output as a confirmed character string (converted character string) to an application or the like (word processor software or the like) (step S8). In this embodiment, the calibration support information is not output to an application or the like. However, since the calibration support information may be useful on the application side, it may be output after clarifying that it is the calibration support information. In particular, since the proofreading support information is attached, it is possible for the application side to determine that the user wanted to use such an expression even against the displayed proofreading support.

In step S8, the display color on the screen is changed to clarify that the character string is confirmed. Also, the display of the calibration support information is stopped.

If one sentence is over when the confirmation command is given and a confirmed character string is output (the sentence ends with “.”), The contents of the phrase candidate storage unit 6 in FIG. 7 are cleared. When a confirmation command is issued in the middle of a sentence, the contents of the phrase candidate storage unit 6 are held as they are. That is, in this embodiment, the contents of the phrase candidate storage unit 6 are held until all the phrases of one sentence are confirmed.

-Processing of continuous judgment of particle-
Next, the process for determining whether or not calibration support is necessary in step S4 will be described. In this embodiment, the judgment in step S4 is performed for continuation of particles, continuation of modifiers, style, passive expression, and the like. First, particle continuation determination processing will be described with reference to the flowchart of FIG.

The continuation of particles here means that the particles “no” and “ga” are consecutive, such as “My child's problem…” and “Hurry up on the bicycle…”. The case where it is highly likely that In such a case, as shown in FIG. 10A, a proofreading support of {continuous particle} is displayed to give a notice to that effect to the user. It is a clause that includes noun equivalents (general nouns, proper nouns, nouns varieties, numbers, suffixes, classifiers, post-conversion words), and the particles at the end of the adjuncts are “no” “de” “ga”. Judgment is made on what is "O".

First, in step S20, the CPU 22 stores in the memory 28 a counter N for “no”, a counter D for “de”, a counter G for “ga”, and a counter WO for “o”. Secure and clear the contents. Next, it is determined whether or not the first phrase (“I” in FIG. 10A) of the converted character string candidate includes a noun equivalent (step S21). This determination is made by referring to the stored contents of the phrase candidate storage unit 6 (see FIG. 7). For example, since “me” is stored as “general noun, case particle”, it is determined that the phrase includes a noun equivalent word.

If the phrase includes a noun equivalent word, it is determined whether or not the phrase includes an appendix (step S22). As described above, with regard to “my”, the “case particle” that is an appendix is stored in the phrase candidate storage unit 6, and therefore, it is determined that the appendix includes the appendix. Further, it is determined whether or not the end of the attached word is a particle (step S23). Here, it is determined to be a particle.

If it is determined that the particle is a particle, the type of the particle is examined (step S24). Here, since it is “no”, the process proceeds to step S25, and the counter N is incremented. Next, it is determined whether or not the counter N is “1” (step S26). Since the counter N is “1” at the beginning of the sequence of “no”, the process returns to step S21 via steps S27, S28, S29, S30, and S33. That is, the above process is repeated for the next phrase “child”. Thereby, when the particle “no” continues, the number of times can be stored in the counter N.

It should be noted that in the step S24, the flowchart when “de”, “g” and “m” are omitted. In these cases, the same process as “no” is performed.

For example, in the case of the conversion character string candidate as shown in FIG. 10A, the value of the N counter becomes “3” when the processing of the last phrase is finished. Thereafter, when it is determined in step S30 that there is no clause after this, the process proceeds to step S31. In step S31, it is determined whether any of the counters N, G, D, and WO is equal to or greater than a predetermined value. In this embodiment, the predetermined values for counters N and G are “3”, and the predetermined values for counters D and WO are “2”. That is, the calibration support is performed by repeating three times or more for “no” and “ga” and repeating two or more times for “de” and “wo”. Here, since the counter N is “3” and is equal to or greater than a predetermined value, a flag indicating that the particle continues is set for the current phrase. That is, as shown in FIG. 11A, the continuous particle item is set to “1” in the column of the phrase number 3 in the flag table.

In step S5 of FIG. 5, based on this flag, the proofreading support information {continuous particle} is displayed after the third phrase (see FIG. 10A).

In the process of continuously determining particles in FIG. 9, when different particles continue, the counter is cleared (step S29). Therefore, as shown in FIG. 10B, when the particle “I” is inserted in the middle, even if there are three particles “NO” as a whole, the proofreading support information of {continuous particle} is not output. However, as illustrated in FIG. 10C, when there is a noun equivalent “immediate” having no attached word, the counter is not cleared and the count is continued. That is, counter clearing is not provided in the processing paths of steps S22, S30, S33, and S21.

Further, in this embodiment, as shown in FIG. 10D, the proofreading support information is displayed at the end of the continuous phrase of the particle rather than being displayed immediately after the phrase exceeding the predetermined value.

Since the flag table as shown in FIG. 11A is provided, the proofreading support information can be displayed even in the middle of the converted character string candidate as shown in FIG. 10E. Although omitted in the flowchart, as shown in FIG. 10F, the counter is cleared when there are punctuation marks or symbols in the middle.

Further, in step S23, even if the particles "~ dano", "~ mono", "~ no de", "~ yo de", and "~ so de" come to the end, the process branches to step S35. This is also omitted in the flowchart.

As is clear from the overall flowchart of FIG. 5, when there is a change in candidate or a change in phrase break, the necessity for proofreading support is determined again (step S4). For example, in the state shown in FIG. 10A, when “child” is changed to “this” which is a conjunction, the display of the calibration support information disappears as shown in FIG. 10G. Similarly, if “my” is changed to “watashino” by post-conversion, the part of speech is only a noun, so the display of the proofreading support information disappears as shown in FIG. 10H.

The phrase candidate storage unit 6 holds part-of-speech information and the like until one sentence is finalized. Therefore, the above-described processing can be performed not only for an undetermined phrase but also for an already confirmed phrase. For example, in FIG. 10A, even when “my” and “child” are confirmed and only “problem” is not confirmed, the calibration support information is displayed in the same manner. Therefore, calibration support with higher practicality can be performed.

-Continuous judgment of modifiers-
Next, the modifier continuous determination process will be described. Here, a series of modifiers means that if there are three or more consecutive clauses of a combination modifier, for example, a sentence such as “white / large / cage / inside ...” This is a case where there is a high possibility that it will be unclear. In such a case, as shown in FIG. 13A, a proofreading support message of {continuation of modifiers} is displayed to give a notice to that effect to the user.

With reference to the flowchart of FIG. 12, the process of a modifier continuous determination is demonstrated. First, in step S40, a plurality of clauses that are combined in the order of a noun equivalent word and a particle equivalent word (according to, in terms of to) are searched for from the converted character string candidates. Such a plurality of clauses is regarded as a clause of one linkage modifying usage (step S40). This determination can be made based on the stored contents of the phrase candidate storage unit 6. For example, the sentence “patent” is composed of a phrase composed of a combination of a general noun “case of patent” and a case particle, and a phrase composed of a combination of a verb stem “used” and an inflection ending. In step S40, the two clauses are regarded as one linkage-modifying usage clause. This is because it can be regarded as a combination of a noun equivalent of “patent” and a particle equivalent of “about”. However, only one phrase is considered in the modifier continuous determination process.

Next, the CPU 22 secures a counter R for the modifier in the memory 28 and clears the contents (step S41). First, it is determined whether or not the part-of-speech of the first phrase is a conjunction (any, great, bad, etc.) (step S42). This determination can also be made by referring to the “part of speech” column of the phrase candidate storage unit 6 (see FIG. 7). If it is a conjunction, the counter R is incremented (step S49). Then, the next phrase is analyzed.

If it is not a conjunction, it is determined whether or not the clause is a conjunction of a predicate (verb, adjective, adjective verb) (step S43). It can be determined by referring to the “part of speech” column of the phrase candidate storage unit 6 and whether it is a combined form or not by referring to the “utilization” column. If it is a continuation form (for example, white, energetic, large, etc.), the counter R is incremented (step S49), and the next phrase is analyzed.

If it is not the union form of the predicate, it is determined whether or not the phrase corresponds to the union usage determined in step S40 (step S44). If the usage is a combination modification, the counter R is incremented (step S49), and the next phrase is analyzed.

If it is not the combined modification usage, it is determined whether or not the phrase is composed of a combination of a noun and a case particle “no” (step S45). If not (that is, if it is not any of the linkage-modifying usages in steps S42 to S45), the counter R determines the number of consecutive linkage-modifying usages (step S46). If the counter R is “3” or more, a modifier continuous flag is set for the phrase immediately preceding the current phrase (see FIG. 11).

If the last phrase is reached in step S52, it is similarly determined whether the value of the counter R is “3” or more (step S54). If the number is 3 or more, the modifier continuous flag in FIG. 11 is set for the current phrase. For example, in the case of a conversion candidate character string as shown in FIG. 13A, a flag is set for the fourth phrase as shown in FIG. 11B.

In step S5 in FIG. 5, based on this flag, proofreading support information {continuation of modifiers} is displayed after the fourth clause (see FIG. 13A).

In this embodiment, the continuation of modifiers is not started by a clause composed of a combination of a noun and “no”. That is, when the counter R is “0”, the counter R is not incremented (step S50). Therefore, the proofreading support information is not displayed for the conversion candidate character string as shown in FIG. 13B. However, once the continuation of modifiers starts, the counter R is incremented due to the presence of a clause with a combination of a noun and “no” (steps S50 and S51). Therefore, the proofreading support information is displayed for the conversion candidate character string as shown in FIG. 13A.

In the modifier continuous determination process of FIG. 12, if there is a clause that is not a clause of the combined modifier usage, the counter R is cleared (steps S45, S46, S48, S, 49, S41). ). Therefore, as shown in FIG. 13C, if there is a phrase “in” in the middle, the proofreading support information {continuation of modifiers} will not be output even if there are three phrases in the combined modification usage as a whole. However, as shown in FIG. 13D, when there is a noun equivalent “bamboo” having no attached word, the counter R is not cleared and the count is continued. Therefore, the calibration support information is displayed. This process is omitted in the flowchart of FIG.

Further, in this embodiment, as shown in FIG. 13A, instead of displaying immediately after the phrase “cage” exceeding the predetermined value, the last phrase “in” in the consecutive linkage-modifying usage phrase The calibration support information is displayed later.

Since the flag table as shown in FIG. 11B is provided, the proofreading support information can be displayed even in the middle of the converted character string candidate as shown in FIG. 13E. Although omitted in the flowchart, as shown in FIG. 13F, the counter is cleared when there are punctuation marks or symbols in the middle.

As is clear from the overall flowchart of FIG. 5, when there is a change in candidate or a change in phrase break, the necessity for proofreading support is determined again (step S4). This is the same as in the case of continuous particles.

The phrase candidate storage unit 6 holds part-of-speech information and the like until one sentence is finalized. Therefore, the above-described processing can be performed not only for an undetermined phrase but also for an already confirmed phrase. This is also the same as the case of particle continuation.

In the above-described embodiment, what satisfies the condition shown in any of Steps S42, S43, S44, and S45 is a clause of a combined modification usage. However, from the viewpoint of the processing speed and the degree of required calibration support, any of them may be treated as not being a combined modification usage. In addition, the conditions may be relaxed or tightened.

-Judgment of style-
Next, the style determination process will be described. Here, the stylistic means whether it is a delicate expression (respected body, more and more keys) or a normal expression (normal body, more keys). In this embodiment, if the user specifies a style in advance, and the style of the converted character string candidate is different from that of the conversion character string candidate, a calibration support display of {a key} or {mass key} is performed, The user is warned to that effect.

The style determination process will be described with reference to the flowchart of FIG. First, in step S60, the CPU 22 sets a honorific flag K and a normal flag J in the memory 28, and clears the contents. Next, it is determined whether or not the first phrase of the conversion character string candidate includes any of the auxiliary verbs “is”, “mass”, and “we are” (including all the utilization forms) (step S61). This can be determined based on the stored contents of the phrase candidate storage unit 6. If it is included, it is determined to be a honorable body, and the honorable body flag K is set to “1” (step S65). For example, as shown in FIG. 15A, when the conversion character string candidate is “I am a cat”, the first phrase “I'm a cat” does not include the auxiliary verb. Accordingly, the process proceeds to step S62.

In step S62, it is determined whether or not the clause includes the imperative command forms "please", "please", and "please" (La line special). This determination can also be made with reference to the “part of speech” and “utilization” columns of the phrase candidate storage unit 6. If it is included, it is determined to be a honorable body, and the honorable body flag K is set to “1” (step S65). In the example of FIG. 15A, the first clause “student is” does not include the above-described command form, so the process proceeds to step S63.

In step S63, it is determined whether or not the clause is immediately before a punctuation mark or symbol. Since “student” does not correspond to this, after steps S63 and S72, step S61 and subsequent steps are repeated for the next phrase.

The next phrase “is a cat” does not correspond to any of steps S61 and S62. Since “is a cat” is a phrase immediately before a punctuation mark, the process branches to step S64 in step S63. In step S64, it is determined whether or not the style is normal. In this way, the determination as to whether or not the object is normal is made based on the phrase immediately before the punctuation mark or symbol.

In this embodiment, it is determined that the object is normal when any of the following conditions is satisfied.

a1) When the phrase does not include a predicate and is a body + final particle "ka": Example "Is it a problem?"
a2) When the phrase does not include a predicate and is a body + "no" + final particle b1) When the last word or auxiliary verb is imperative: Example "Think."
c1) When the last sentence or auxiliary verb of the clause is a final form + connected particle: Example “Thinking,”
c2) When the last phrase or auxiliary verb is a final form + final particle c3) When the last phrase or auxiliary verb is a combined + connected particle c4) The last phrase or auxiliary verb is a final form + final particle Case d1) When the last sentence of the clause is a combination form + "no" + final particle: Example "Do you think?"

Here, “being a cat” corresponds to the above, and is thus determined to be normal. Therefore, the normal flag J is set to “1” (step S71).

The next phrase is “.”, Which ends the sentence, and branches from step S66 to S67. In step S67, it is determined whether the preset style (stored in the memory 28 with an area) is different from the determined style. If the setting style is “respect style”, it is different from the judgment style, so a style error display flag is set (step S68).

In step S5 in FIG. 5, based on this flag, proofreading support information {key to be} is displayed after the last phrase (see FIG. 15B).

In the above case, if the setting style is “normal”, it matches the determination style, so the proofreading support information is not displayed as shown in FIG. 15A.

If the setting style is “normal” and the conversion character string candidate is as shown in FIG.

Further, in step S69, it is determined whether or not the flags K and J are both “1”. The fact that both flags are “1” indicates that the style has changed in one sentence. For example, in the case of a converted character string candidate as shown in FIG. 15D, both flags are “1”. In this case, a style change display flag is set in step S70.

In step S5 of FIG. 5, based on this flag, proofreading support information {changed style} is displayed after the last phrase (see FIG. 15D).

In this embodiment, a change in style is determined for one sentence. However, if the previous sentence that has been finalized to the last phrase is stored in a buffer or the like, it is possible to determine the style change over a plurality of sentences.

As is clear from the overall flowchart of FIG. 5, when there is a change in candidate or a change in phrase break, the necessity for proofreading support is determined again (step S4). This is the same as in the case of continuous particles.

The phrase candidate storage unit 6 holds part-of-speech information and the like until one sentence is finalized. Therefore, the above-described processing can be performed not only for an undetermined phrase but also for an already confirmed phrase. This is also the same as the case of particle continuation.

In the above-described embodiment, the one satisfying the conditions shown in any one of the aforementioned a1) a2) b1) c1) c2) c3) c4) d1) is used as a normal. However, from the viewpoint of the processing speed and the degree of required calibration support, any of them may be treated as not a normal one. In addition, the conditions may be relaxed or tightened. The same is true for respect.

-Judgment of passive expression-
Next, the passive expression determination process will be described. When a phrase with passive expression is found, the proofreading support information {passive expression} is displayed to alert the user. This determination is made based on whether or not there is a phrase including the auxiliary verbs “re” and “re”. For example, as shown in FIG. 16A, in the case of the conversion character string candidate “it seems to be like this”, the phrase “it seems” is detected, and the passive display flag becomes “1”. In response to this, a calibration support display is made as shown in FIG. 16A.

-Judgment of unrepresented expression-
Next, the process of determining the blank expression will be described. In this embodiment, when a verb corresponding to a free expression is registered in an independent word dictionary, the fact that it is a “free expression” is stored as proofreading related information. As shown in FIG. 3, “removed expression” is stored as proofreading-related information for “eaten”. The CPU 22 stores this proofreading related information in the phrase candidate storage unit 6 when generating a phrase in step S2 of FIG. Further, when displaying in steps S5 and S6, if there is proofreading related information, it is displayed after the phrase (see FIG. 16B).

Note that the proofreading related information is not limited to “rough expression” but can also store other information useful for proofreading.

-Display of calibration support information-
In each of the above embodiments, the calibration support display is performed on the same line as the target conversion character string candidate. However, other display methods that can be distinguished from conversion character string candidates may be used. For example, the calibration support information may be displayed in the balloon 100 as shown in FIG. In this way, the proofreading support information can be displayed on different lines, so that the readability of the input sentence is not impaired by the proofreading support information. In addition, the relationship with the phrase for which the proofreading support information should be displayed does not become unclear. However, the characters in the line near the input line are hidden, making it difficult to read.

Further, the proofreading support display area 102 may be provided so that display is performed regardless of the position of the input character. This method does not have the above-mentioned drawbacks, but it is difficult to confirm the presence or absence of calibration support information.

Moreover, you may display as shown in FIG. That is, when the cursor CK of the target phrase is not in the phrase for which the calibration support display is to be performed, a mark such as * is displayed in the corresponding phrase (see FIG. 18A). When the cursor CK moves to the target phrase, calibration support information is displayed as shown in FIG. 18B. Further, instead of moving the attention phrase cursor, it may be displayed by double-clicking * with a mouse or the like.

In addition to display on the screen, information may be output by voice or the like. Furthermore, the data may be output to other software or the like.

Each of the above display methods may be used alone or in combination of two or more methods.

-Others-
In each of the above embodiments, only the calibration support information is displayed. However, calibration candidates may be displayed, and this may be selected by the user for correction input.

Further, whether or not to perform each of the above-described calibration supports can be selected by the user for each calibration support.

In each of the above embodiments, each function of FIG. 1 is realized using a CPU, but a part or all of the functions may be configured by hardware logic.

Description of figure

It is a figure which shows the whole structure of the character string converter with a proofreading assistance function by one Embodiment of this invention. It is a figure which shows the hardware constitutions at the time of implement | achieving the character string converter of FIG. 1 using CPU. It is a figure which shows the data structure of the independent word dictionary. 3 is a diagram showing a data configuration of an attached word dictionary 10. FIG. It is a flowchart which shows the whole conversion process and proofreading assistance process. It is a figure for showing a phrase generation process notionally. FIG. 6 is a diagram showing the contents stored in a phrase candidate storage unit 6. It is a figure which shows the character string conversion candidate in a conversion process. It is a flowchart of the continuous determination of a particle. It is a figure which shows the form which displays the continuation of a particle as proofreading assistance information. It is a figure which shows a flag table. It is a flowchart of the continuous determination of a modifier. It is a figure which shows the form which displays the continuation of a modifier as proofreading assistance information. It is a flowchart of style determination. It is a figure which shows the form which displays the correction assistance information regarding a style. It is a figure which shows the display of other proofreading assistance information. It is a figure which shows the display method of calibration assistance information. It is a figure which shows the other display method of proofreading assistance information.

Explanation of symbols

2 ... Conversion means 4 ... Kana character string storage unit 6 ... Phrase candidate storage unit 8 ... Autonomous word dictionary 10 ... Attached word dictionary 14 ... Phrase generation means 16 ... Phrase candidate Selection means 18 ... composition calibration support means 20 ... display means

Claims (24)

In a character string conversion device that converts a given kana character string into a conversion character string including kanji,
A dictionary that stores the converted character string and its part of speech in association with the kana character string;
With reference to the dictionary, a conversion character string candidate obtained by dividing a given kana character string into a conversion character string by dividing it into clauses is output, and upon receiving a confirmation command, the selected conversion character string candidate is determined as a confirmation character Conversion means for outputting as a column;
Proofreading support means for performing output for proofreading support based on the part of speech of each phrase in the conversion character string candidate,
A character string conversion device comprising:   2. The character string conversion device according to claim 1, wherein the proofreading support means determines whether or not an adjunct in each phrase in the converted character string candidate is a particle, and a phrase having the same particle continues for a predetermined number of times. , Proofreading output indicating that particles are continuous.   2. The character string conversion device according to claim 1, wherein the proofreading support means determines whether or not each phrase is in a combination modification usage in a conversion character string candidate, and the phrase of the combination modification usage continues for a predetermined number of times. In such a case, a proofreading support output indicating that the modifiers are continuous is performed.   4. The character string conversion apparatus according to claim 3, wherein the proofreading support means finds a phrase composed of any one of a) a conjunction, b) a conjunction form of a predicate, c) a noun equivalent and a particle equivalent. , It is judged that the application method of the combination modification is continued once by this clause, and after this clause, a) a conjunction, b) a conjunction form of a predicate, c) a combination of a noun equivalent and a particle equivalent, d) a noun The continuation of the clause of the combination modification usage is determined based on how many times the clause constituted by any combination of the particle “no” continues.   2. The character string conversion apparatus according to claim 1, wherein the proofreading support means determines whether the sentence is normal or respected based on the part of speech of the phrase in the converted character string candidate, and proofreads based on the determination. It is characterized by performing support output.   6. The character string conversion device according to claim 5, wherein the proofreading support means is that the conversion character string candidate clause is “is”, “mas”, “has”, or other utilization forms of these, or “please”. It is characterized by judging that it is a respectful body when it includes “please” and “speak”. 6. The character string conversion device according to claim 5, wherein the proofreading support means determines that the conversion character string candidate is normal when the phrase immediately preceding the punctuation mark or symbol is one of the following: It is characterized by that.
a) Combination of the body and final particle "ka" b) Combination of body and "no" and final particle c) When the last phrase or auxiliary verb is imperative d) The last sentence of the phrase is connected to the final form In the case of a combination of particles, e) In the case where the last sentence of the clause is a combination of a final form and a final particle, f) In the case where the last sentence of the phrase is a combination of a combined form and a connected particle, g) In the case of a combination of final form and final particle h) In the case where the last sentence of the phrase is a combination of the combined form, “no”, and final particle.   8. The character string converter according to claim 5, 6 or 7, wherein when the predetermined “normal” or “respectful” style and the style determined by the proofreading support means are different, the style is different. It is characterized by outputting the effect.   8. The character string conversion apparatus according to claim 5, wherein the proofreading support means determines that the style of the sentence has changed when it is determined that the sentence style of each phrase is not the same for one sentence of the converted character string. It is characterized by performing calibration support output.   2. The character string conversion apparatus according to claim 1, wherein the proofreading support means outputs a proofreading support output indicating that the phrase is a passive expression when the phrase includes the auxiliary verbs “de” and “being”. In a character string conversion device that converts a given kana character string into a conversion character string including kanji,
A dictionary storing conversion character strings and proofreading related information in association with kana character strings;
With reference to the dictionary, a conversion character string candidate obtained by dividing a given kana character string into a conversion character string by dividing it into clauses is output, and upon receiving a confirmation command, the selected conversion character string candidate is determined as a confirmation character Conversion means for outputting as a column;
The conversion means is proofreading support means for performing an output for proofreading support based on the proofreading related information obtained from the dictionary in association with each phrase;
A character string conversion device comprising: In a character string conversion device that converts a given kana character string into a conversion character string including kanji,
A character string conversion apparatus characterized in that a conversion character string candidate for a given kana character string is displayed on a display screen, and proofreading support information for the conversion character string candidate is also displayed or output by voice.   13. The character string conversion device according to claim 12, wherein the proofreading support information is displayed in a line near the conversion character string candidate and different from the conversion character string candidate.   13. The character string conversion device according to claim 12, wherein the proofreading support information is displayed on the same line as the character string candidate.   13. The character string conversion device according to claim 12, wherein the proofreading support information is displayed at a predetermined position irrespective of a predetermined display position of the character string candidate.   In the character string conversion device according to claim 12, the conversion character string candidate for the given character string is displayed on the display screen, and a display indicating that there is proofreading support information for each clause of the conversion character string candidate, The proofreading support information is displayed only when the cursor is moved to a phrase that is displayed with proofreading support information.   A conversion character string in which a conversion character string and its part of speech are stored in a dictionary in advance in association with a kana character string, and the given kana character string is divided into phrases and converted into a conversion character string with reference to the dictionary In the character string conversion method for outputting a candidate and receiving a confirmation command and outputting the selected converted character string candidate as a confirmed character string, based on the part of speech of each clause in the converted character string candidate, A character string conversion method characterized in that output is performed.   18. The character string conversion method according to claim 17, wherein it is determined whether or not an adjunct in each phrase in the converted character string candidate is a particle, and if a phrase having the same particle continues for a predetermined number of times, the particle continues. It is intended to perform calibration support output to the effect.   18. The character string conversion method according to claim 17, wherein, in the conversion character string candidate, it is determined whether or not each clause is in a combination modification usage, and if the clause in the combination modification usage continues for a predetermined number of times, the modifier is It is intended to output calibration support output indicating that it is continuous.   18. The character string conversion method according to claim 17, wherein it is determined whether the sentence is normal or respected based on the part of speech of the phrase in the conversion character string candidate, and proofreading support output is performed based on the determination. It is characterized by being.   18. The character string conversion method according to claim 17, wherein when the phrase in the conversion character string candidate includes the auxiliary verbs “re” and “required”, a proofreading support output indicating passive expression is performed.   The conversion character string and the proofreading related information associated with the kana character string are stored in a dictionary in advance, and the conversion character string candidate obtained by converting the kana character string into a conversion character string by dividing the phrase into phrases is referred to the dictionary. In the character string conversion method that outputs the selected conversion character string candidate as a confirmed character string in response to the confirmation command, the calibration support information is obtained based on the proofreading related information obtained from the dictionary in association with each phrase. A character string conversion method characterized in that output is performed for the purpose.   In a character string conversion method for converting a given kana character string into a converted character string including kanji, a conversion character string candidate for the given kana character string is displayed on the display screen, and proofreading support information for the converted character string candidate The character string conversion method characterized by outputting in a display or audio | voice together. A computer-readable storage medium storing a computer-executable program, wherein the program is for realizing the apparatus or method of any one of claims 1 to 22 using a computer. Features

Images