<?xml version="1.0" standalone="yes"?>
<Paper uid="C00-1050">
  <Title>Kana-Kanji Conversion System with Input Support Based on Prediction</Title>
  <Section position="3" start_page="0" end_page="341" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> TOSHIBA developed the world's first Japanese word processor in 1978. Unlike languages based on an alphabet, Japanese uses /,housands of Ica nji characters of varying comp\]exity. Hence, l,o arrange all of l~a'~:ii chm'acl;ers on keyboard is; difficult. On the other hand, kana dlaracters which are phonetic scripl,s of Japanese have 83 variations; these can be arranged on keyboa'&amp; As a result, conversion from kana notations to kanji ones, whal, is called kana-ka,~:ji conversion, has been used. Since Japanese is not writl,en separately by words, segmental,ion of typed kana characl,er strings has ambiguil,y.</Paragraph>
    <Paragraph position="1"> And ~m ambiguil,y in conversion exists, too; a kana notation may correspond l,o some different t:a, nji notations. These make t,:ana-kanji conversion challenging.</Paragraph>
    <Paragraph position="2"> We have made efforts l,o raise a precision of kana-kaTt:ji conversion, thinking l,h~t high precision can provide a better input enviromnent for the user. A t)rccision of our kana-kanji conversion system reaches 95-98% for several kinds o\[' texts in our previous experiments. Neverthe\]ess, this approach is not; enough in the situations where fasl; typing is hard, e.g., lbr beginners who m'e not familiar wil,h keyboard or for palm-size comlmters. Thus, new method to reduce key input; operations is needed.</Paragraph>
    <Paragraph position="3"> We propose a k(vna-kanji conversion system witll input support based on prediction. This system is composed o\[' two parts: prediction of succeeding ka'nji character strings t\]'om typed kana ones, and ordinary hrvna-kanji conversion.</Paragraph>
    <Paragraph position="4"> It; automatically shows candidates of ka'nji character strings which the user intends to input.</Paragraph>
    <Paragraph position="5"> The candidates change as /;lie usc, r inputs ka.na characters. If no approt)riate ctloice is presenl;e(t, the candidates aul;Olnatictd\]y disappear when the next kana character is entered. Our system, l,here\['ore, can be used in the same manner as an ordinary ka'na-kanji conversion system, and allows the user to save time and efforts for key input without learning new key operations. null We have been considered two issues to generate accurate candidates: (i) How Co determine where t;yped kana character strings are segmented; since Japanese is not written separately by words, determination of positions where words start is needed to retrieve dictionaries.</Paragraph>
    <Paragraph position="6"> (ii) How to determine when prediction candidates are presented; if all of retrieval results are always shown, a sysl,em cannot be convenielfl,.</Paragraph>
    <Paragraph position="7"> Surveying previous works Dora the view on above issues, we found l,hal, the Reacl,ive Keyboard has been proposed (Darragh el, al., 1980).</Paragraph>
    <Paragraph position="8"> It accelerates typewritten communication with  a computer system by predicting what the user is going to type next. In this system, the top of typed ct~aracter strings is regarded as the top of words, because English is written separately by words; the issue of segmentation of character strings does not occur.</Paragraph>
    <Paragraph position="9"> On the other hand, in previous works for Japanese, a predictive pen-based text input method has been proposed (Fukushima and Yalnada, 1.996). In this system, character strings are input by h~md-writing on LCD panel. Since the user usually inputs not only by kana but also by t~anji and an alphabet, entered character strings are segmented with the help of the wu-iety of characters. Thus, the issue of segmentation is not considered.</Paragraph>
    <Paragraph position="10"> The POBox (Pen-Operations Bused On eXample) which is a text input method tbr pen-based computers has also been proposed (Masui, 1998). It shows succeeding candidates from cl~aracter strings input by software keyboard.</Paragraph>
    <Paragraph position="11"> Arbitrary positions of input character strings can be regarded as the top of words, and retrieval results are always shown as candidates; the prediction ordering is based on the user's previous choice. Since input speed by pen is not faster than that by keyboard, time to choose candidates is shorter than that to input c.haracters. Hence, even if many candidates are shown, this n~ethod is effective for pen-based computers. It is, however, inefficient for ordinary keyboard. null We propose a system with following features: null (i) Arbitrary positions of typed kana character strings are regarded as the top of words.</Paragraph>
    <Paragraph position="12"> (ii) A system dictionary and a user dictionary are used, aud each entry in the system dictionary has certainty factor calculated from the frequency of words in corpora.</Paragraph>
    <Paragraph position="13"> (iii) Candidates are estimated by certainty factor and useful'hess factor; and likely ones with greater factors than thresholds are shown.</Paragraph>
    <Paragraph position="14"> These features provide an cfl3cicnt Japanese text input environment for ordinary keyboard without learning new key operations.</Paragraph>
    <Paragraph position="15"> Section 2 shows an example of text input using the proposed system. Section 3 explains an input support method based on prediction. Section 4 shows efficiency of our system by means of experiments. Section 5 describes conclusions.</Paragraph>
  </Section>
class="xml-element"></Paper>