JPH07122874B2 - Homophone learning device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION Industrial Field of the Invention The present invention, when inputting a kana-kanji character as a corresponding kanji-kana mixed notation of an input kana character string, inputs a specific word from a plurality of homophones. This is a homophone learning device for efficient selection.

2. Description of the Related Art In recent years, a kana-kanji conversion device, which has been regarded as a particularly promising device for easily creating Japanese documents by inputting kana characters, is equipped with a homophone learning mechanism for quickly selecting the correct word from homophone variants. Have been developed.

For example, a so-called homophone usage frequency learning device changes the search order designation information of homophones in the word dictionary based on the usage frequency of the words used in the word dictionary, and quickly obtains the frequently used words. Is possible.

Also, a device called a homophone usage order learning device stores the usage history information of the words in order and stores them in the input kana string each time a specific word is selected from the words having a plurality of homophones in the word dictionary. On the other hand, the homophone selected immediately before is used as the first search word, and the homophon selected before that is sequentially output as the second search word, so that the word used recently can be obtained earlier.

Conventionally, when the functions of the homophone usage frequency learning and the homophone usage sequence learning are realized together, a mechanism corresponding to each is provided.

For example, in the homophone usage order learning, the word identifiers that can uniquely limit the words are ordered and stored in the homophone order memory unit, and when the word is searched, the homophone word order memory unit is sequentially searched to search for the homophone words. The order of use learning was realized.

In the homophone usage frequency learning, the usage frequency information in the word dictionary is updated every time a homophone is selected, and the homophone usage frequency learning is performed by searching for the words in the homophony in the order sorted by this usage frequency. Was realized.

By having these two mechanisms, both homophone usage sequence learning having the meaning of short-term learning and homophone frequency learning having the meaning of long-term learning can be performed.

(For example, JP-A-52-132653 and JP-A-55-153033.
However, in such a conventional homophone learning device,
Even if we combine frequency learning and sequential learning, short-term learning still prioritizes sequential learning, so if a specific word that is used significantly in a particular sentence occurs, a short-term learning that prioritizes only this word will occur. It is not possible to do frequency learning.

A conventional homophone learning device will be described below. FIG. 3 is a block diagram showing the configuration of a conventional homophone learning device.

Reference numeral 1 is a kana input section for inputting phonetic character strings such as Japanese hiragana and katakana, a character conversion instruction code z1, and a hiragana display instruction code z2.

Reference numeral 2 is a word dictionary that stores Japanese word readings in association with the corresponding word notations in the order of Japanese syllabary. Reference numeral 3 denotes a character conversion unit for sequentially searching a kana character string input from the kana input unit in a homophone order storage unit 4 and the word dictionary 2 described later to obtain a corresponding word notation. At this time, the homophone word order storage unit 4 is searched in the reverse order of the words stored in the homophone word order storage unit 4. Also, the word dictionary 2
When there are a plurality of words having a pronunciation that matches the input kana string in the search of, the search is performed in the order of frequency, and the word is output to the homophone selection unit 5 described later together with the input kana string.

A homophone word order storage unit 4 stores homophone different characters selected by a homophone word selection unit 5 described later by associating the reading and the notation in the selected order in the first-in first-out system. Reference numeral 5 denotes a homophone selection unit which, when there are a plurality of output word notations from the character conversion unit 3, that is, when there are homophone variants, these are designated as the next candidate instruction code z3 from the homophone designation device 8 described later. By inputting, it is possible to sequentially display on the display unit 6 described later and allow the operator to select a desired notation.

A display unit 6 displays the word notation selected by the word selection unit 4 on the screen.

Reference numeral 7 is a frequency updating unit for adding 1 to the frequency in the word dictionary 2 for the designated word from the homophone selection unit 5.

Reference numeral 8 denotes a homophone designation device, which outputs a next candidate designation code z3 and a definite code z4 meaning completion of selection in a homomorphic character to the homophone selection unit 5.

Problems of this conventional homophone learning device will be described below with reference to FIG. FIGS. 4A and 4B show the contents of the document appearing on the display section during the preparation of the document using the conventional homophone learning device. Further, FIG. 6C shows the contents of the frequency storage unit in the word dictionary 2 for frequency learning. In other words, for those that have not been subjected to order learning, those with a high frequency described in this frequency storage unit are output in order. In this example, “publish”, “sightseeing”, “practice”, “completion”, and “daring” are output in this order.

FIG. 4a shows the display state of the document during creation. At this time, paying attention to the word having the pronunciation "Kankou" stored in the homophonic order storage unit, "homogeneous", "sightseeing", and "custom" have been recorded in the homophoneic order storage unit so far.
The homophones are stored in this order. In ordinary writing, it is often the case that a plurality of types of homophones are selected, but generally, there is considerable variation in the frequency of use among these homophones. In the document shown in FIG. 4, "tourism" is used in preference to "customs".

Now, at this time, when the operator uses the word "sightseeing" again, in the conventional homophone learning device, based on the homophone order learning process that is prioritized in the short-term learning, simply prioritize the immediately preceding homophone. To output, "custom" is selected as the first candidate and displayed on the screen. Inevitably, the operator needs to perform a process of selecting another candidate.

FIG. 4b is a diagram showing the display state of the document at the time when this process is completed.

Problems to be Solved by the Invention That is, generally, each time a document is created, there is a considerable difference in the frequency of use of each homonym, but in a conventional homonym learning device, it appears by chance. Since the homophonic word of is preferentially displayed, there is a problem that the word originally desired to be displayed is postponed.

The present invention has been made in view of the above problems, and with a simple configuration, it is possible to perform short-term usage frequency learning according to the homophone usage status during document creation, and efficiently output prioritized homophones. The purpose of the present invention is to provide a homophone learning device.

Means for Solving the Problems In order to achieve the above-mentioned goal, the present invention provides a kana input section for inputting a Japanese reading string, a word dictionary storing word reading and notation, and a plurality of homophones. A homophone selection unit that selects one from the words, a homophone order storage unit that stores the reading and notation of the word selected by the homophone selection unit in association with the selected order, and the homophone order storage The maximum frequency word selection unit that searches for the maximum number of words with the same reading and the same notation stored in the set, and the reading string that is input from the kana input unit, and the words that the maximum frequency word selection unit searches A character conversion unit for converting a kanji-kana mixed character string based on a homophone order storage unit and a word obtained by searching the word dictionary;
A display unit that displays the kanji-kana mixed character string, and if a homophone exists in the kanji-kana mixed character string, the word searched by the maximum frequency word selection part, the word searched from the homophone order storage part, The homophone learning device is characterized in that the words retrieved from the word dictionary are displayed in order.

Action The present invention, with the above-mentioned configuration, simply uses only the word use information stored in the homophone word order storage unit by the function of the maximum frequency word selection unit capable of counting the number of word duplications in the homophone word order storage unit. Can perform short-term frequency learning. This makes it possible to perform short-term homophone usage frequency learning and homophone usage order learning in combination.

Practical Example FIG. 1 is a block diagram of a homophone learning device in one embodiment of the present invention. Reference numeral 1 is a kana input section for inputting phonetic character strings such as Japanese hiragana and katakana, a character conversion instruction code z1, and a hiragana display code z2.

Reference numeral 2 is a word dictionary that stores Japanese word readings in association with the corresponding word notations in the order of Japanese syllabary. Reference numeral 3 denotes a character conversion unit, which sequentially searches a kana character string input from the kana input unit through a maximum frequency word selection unit 9 described later, a homophone order storage unit 4 described later, and the word dictionary 2 to sequentially display corresponding word expressions. I will get it. At this time, the homophone word order storage unit 4 is searched in the reverse order of the words stored in the homophone word order storage unit 4. Further, in the search of the word dictionary 2, when there are a plurality of words having a reading that matches the input kana string, the words are searched in descending order of frequency and are output to the homophone selection unit 5 to be described later together with the input kana string.

A homophone word order storage unit 4 stores homophone different characters selected by a homophone word selection unit 5 described later by associating the reading and the notation in the selected order in the first-in first-out system. Reference numeral 5 denotes a homophone selection unit, which inputs a next candidate instruction code z3 from a homophone designating device 8 which will be described later when there are a plurality of output word notations from the character conversion unit 3, that is, when there are homophone variants. By this, it is possible to sequentially display on the display unit 6 described later and the operator can select a desired notation.

A display unit 6 displays the word notation selected by the word selection unit 4 on the screen.

Reference numeral 7 is a frequency updating unit for adding 1 to the frequency in the word dictionary 2 for the designated word from the homophone selection unit 5.

Reference numeral 8 denotes a homophone designation device, which outputs a next candidate designation code z3 and a definite code z4 meaning completion of selection in a homomorphic character to the homophone selection device 5. A maximum frequency word selection unit 9 counts the number of homophones in the homophone order storage unit 4, obtains the maximum value, and outputs it to the character conversion unit 3.

The operation of the homophone learning device of the present embodiment configured as described above will be described below with reference to the flowchart of FIG.

〔Character conversion〕

(1) The kana string k entered by the operator from the kana input section 1 is sent to the character conversion section 3 in a format with the conversion instruction code z1 or the hiragana display instruction code z2 added at the end.

(2) When the kana string k input from the kana input unit 1 has a hiragana display instruction code z2 at the end, the character conversion unit 3 outputs the hiragana notation of k as a word notation to the homophone selection unit 5.
That is, the control shifts to (5).

(3) The character conversion unit 3 sequentially searches the maximum frequency word selection unit 9, the homophone word order storage unit 4, and the word dictionary 2 for the input kana string k, and selects a word having a corresponding reading together with the input kana string k. It is sent to the part 5 in the order of retrieval. At this time, if there are a plurality of homophone variants, they are all output to the homophone selection unit 5.

(4) The homophone selection unit 5 first determines whether or not the word received from the character conversion unit 3 includes homophone variants. if,
If included, control is transferred to the process (6). If not, the following process (5) is performed.

(5) The display unit 6 outputs the notation of the word delivered from the homophone selection unit 5 to the display unit 6 and displays it on the screen, and then enters the input waiting state.

(6) The homophone selection unit 5 outputs the first received word notation in the delivered homomorphic characters to the display unit 6 and displays it on the screen, and also provides guidance for the operator to select the homophone on the display screen. After sending out, it will be in input waiting state.

[Homophone selection]

(7) When the operator inputs the next candidate instruction code z3 from the homonym designation device 8 while the guidance for homonym selection is displayed, the homonyms are selected in the same order as they are received from the character conversion unit 3. Display one word notation in order.

(8) When the operator obtains the desired word notation on the screen, the operator inputs the confirmation code z4 from the homophone designation device 8. Then, the homophone selection unit 5 outputs the selected word notation to the homophone order storage unit 4 together with the input kana string k.

(9) The homophone word order storage unit 4 stores the selected homophone different characters in the selected order internally. At this time, the homophone order storage unit 4 stores the latest word position information indicating which is the latest stored word. After storing the word, it waits for input again.

Here, the operation of the homonym learning device will be further described using an input example. FIG. 5a shows only the readings of "Kankou" extracted from the contents of the homophonic word order storage unit 4 shown in FIG. 4a at the time when the input of the example sentence is completed. In addition, FIG. 5b shows the extracted contents of the homophonic word order storage unit 4 just after the example sentence of FIG.

At the time of Fig. 5a, the operator reads "Kankou" to obtain the character string "Sightseeing".
When the corresponding reading is keyed in from the kana input device 1, the character conversion unit 3 first obtains the expression "sightseeing" from the maximum frequency word selection unit 9. This means that “Sightseeing” is 2 out of those that have the reading “Kankou” in the homophone order storage unit 4.
No more than one "practice". If the operator wants to input the "practice" that was input immediately before, the character string "practice" can be obtained as the next candidate by giving a command to the homophone selection unit 5.

In this way, not only can valid words be displayed as the first candidate by prioritizing valid words due to the effect of short-term frequency learning, but the word used immediately before can usually be selected as the second or third candidate. It can be said that it also has the effect of learning.

Further, according to the present embodiment, since the conventional frequency learning is also performed at the same time, if there is no word to be referred to in the homophone order storage unit 4, the homophone is preferentially output based on the long-term use frequency. You can

EFFECTS OF THE INVENTION As is clear from the above description, this homophone learning device is
Short-term homophone frequency learning is applied only to unique words that are used significantly in specific sentences, and after the second candidate, the homophone usage order learning is output according to the selected order. It is possible to realize the above-mentioned functions with a compact configuration without having them and without preparing a special storage device for frequency learning.

Further, as shown in the embodiment, since the method of "priority and outputting the word most frequently used recently" is selected, the desired homophone can be displayed with high accuracy.

[Brief description of drawings]

FIG. 1 is a block diagram of a homophone learning device in one embodiment of the present invention, FIG. 2 is a flowchart showing the operation of the homophone learning device in one embodiment of the present invention, and FIG. 3 is a conventional homophone learning device. 4 is a block diagram of FIG. 4, FIGS. 4A and 4B are diagrams showing a display state of the display unit during document creation, FIG. 4C is a diagram in which frequencies of homonyms in the word dictionary are extracted, and FIGS. 5A and 5B.
FIG. 6 is a diagram showing extracted contents of a homophone order storage unit during document creation. 1 ... Kana input section, 2 ... word dictionary, 3 ... character conversion section, 4 ... homophone order storage section, 5 ... homophone selection section, 6
...... Display unit, 7 ... Frequency updating unit, 8 ... Homophone selection instruction device, 9 ... Maximum frequency word selecting unit.

Claims (1)

[Claims] 1. A kana input section for inputting a Japanese reading string,
A word dictionary that stores the reading and notation of words, a homophone selection unit that selects one from a plurality of homophones, and a reading and notation of the words selected by the homophone selection unit, in the selected order. Input from the kana input section, and a homophone word order storage section that stores the word in association with each other, a maximum frequency word selection section that searches for the maximum number of words of the same reading and the same notation stored in the homophone word order storage section, A character conversion unit for converting the read string based on the words searched by the maximum frequency word selection unit and the words obtained by searching the homophone order storage unit and the word dictionary, into a kanji-kana mixed character string. And a display unit for displaying the kanji-kana mixed character string, and when a homophone exists in the kanji-kana mixed character string, the word searched by the maximum frequency word selection part, the homophone word order storage part are searched. word,
A homophone learning device, wherein the words searched from the word dictionary are displayed in order.