<?xml version="1.0" standalone="yes"?>
<Paper uid="C02-1149">
  <Title>Entering Text with A Four-Button Device</Title>
  <Section position="3" start_page="0" end_page="0" type="metho">
    <SectionTitle>
2 An Example
</SectionTitle>
    <Paragraph position="0"> Figure 1-1 shows the GUI for the TouchMeKey4 keypad. Nine buttons are visible, with four on either side of the central boxes plus a `quit' button on the right-hand side. In this paper, we only count those buttons that are only used for the entry of characters that is, the four on the right-hand side. We also impose the constraint that the buttons may only be pressed one at a time, because the inclusion of key-chords increases the actual number of buttons by including the combinations of keys.</Paragraph>
    <Paragraph position="1"> Six or seven letters of the alphabet are assigned to each of the buttons. The no. 1key has `abcdef', the no. 2 key has `ghijkl', the no. 3key has `mnopqrs', and the no. 4 key has `tuvwxyz'. The small letters</Paragraph>
  </Section>
  <Section position="4" start_page="0" end_page="2" type="metho">
    <SectionTitle>
TouchMeKey4 keypad
</SectionTitle>
    <Paragraph position="0"> are assigned to the same keys as the corresponding capital letters. All other ASCII characters other than the alphanumeric characters are assigned to the no.</Paragraph>
    <Paragraph position="1"> 4key.</Paragraph>
    <Paragraph position="2"> Suppose that wehavejustentered the string `human language'. The text appears in the upper box in the middle of the window (the upper text-box in Figure 1-1). We now wish to enter the word `technology'. Words are entered through a single-tapper-character form of predictiveentry;; a key is only pressed once to enter a character. For example, the no. 4 button is pressed once to enter the `t' of `technology'. Toenter the subsequent `e', the no. 1 button is pressed once.</Paragraph>
    <Paragraph position="3"> After the no. 1 button has been pressed, the Touch-</Paragraph>
    <Paragraph position="5"> see two dierences from Figure 1-1. The rst is that `41' appears in the box in the middle of the window.</Paragraph>
    <Paragraph position="6"> This indicates the string that the user has just entered. The second change is that some words have appeared in the lower box in the middle of the window (a list-box that we call the `candidate-box'). These words are the candidate words that correspond to the user's input, `41'.</Paragraph>
    <Paragraph position="7"> Eachpressof abutton bythe usermakesthe TouchMeKey4 system automatically search the dictionary for candidates. The candidates include longer words as well as, if such words exist, words of the same length as the entered sequence of digits. The candidates are thus all words that begin with one letter from `tuvwxyz' followed by one letter from `abcdef'.</Paragraph>
    <Paragraph position="8"> For example, `text', `was', and `vendors' are candidates, as is the two-character candidate `we'.</Paragraph>
    <Paragraph position="9"> The numerous candidates are sorted into an order before they areplaced in the candidate box and shown to the user. The order is according to word probability as determined on the basis of PPM (prediction by partial match), which has been proposed in the information-theory domain. A detailed description is given in x4, but we summarize the method's essence here as part of our explanation of Figure 1. The relevance of each candidate is measured by statistics from two sources.</Paragraph>
    <Paragraph position="10"> Base dictionary the unigram statistics collected fromahuge corpus of newspaper data, and User corpus the ngram statistics obtained from a small personal document supplied by the user.</Paragraph>
    <Paragraph position="11"> In this example, the Base dictionary is constructed from one year of issues of the Wall Street Journal (WSJ) that contains 93 000 dierent words and the User Corpus is a computer magazine that contains 10 000 words. The particular User corpus is the reason for the appearance of the relatively uncommon word `vendors' among the top ve candidates (Figure 1-2).</Paragraph>
    <Paragraph position="12"> Our target `technology' appears as the secondranked candidate. In selecting this word, the user highlights it by using the down button on the left- null Note that the most recently pressed button is framed bya thick line.</Paragraph>
    <Paragraph position="13"> hand side of the window(Figure 1-3) and then presses the enter button (Figure 1-4). We see that the selected candidate now appears in the upper text-box  .</Paragraph>
    <Paragraph position="14"> In describing our realization of the TouchMeKey4 system outlined above, the following four questions are discussed in the remainder of this paper: Interface Is some method other than that described above suitable for text entry with a four-button device? Candidate Estimation How can the system estimate the relevance of each candidate? Key Assignment How should characters be assigned to the individual buttons? Number Of Keys What is the minimum numberof keys required? Is the entry of free text with only two buttons reasonably ecient?</Paragraph>
  </Section>
  <Section position="5" start_page="2" end_page="2" type="metho">
    <SectionTitle>
3 Interface
</SectionTitle>
    <Paragraph position="0"> Various methods for the entry of text via a four-button device are conceivable. The biggest choice is whether or not to adopt a predictive method.</Paragraph>
    <Section position="1" start_page="2" end_page="2" type="sub_section">
      <SectionTitle>
3.1 Non-PredictiveEntry Methods
</SectionTitle>
      <Paragraph position="0"> Let's start by considering the case where we don't adopt prediction. This means that we need to enable the exact entry of the individual characters via the four buttons. One method of this type involves assigning an order to the characters on eachkey;;akey is then pressed i times to obtain the i-th character (we call this the multi-tap method). This method is commonly applied on mobile phones.</Paragraph>
      <Paragraph position="1"> However, there are two problems with this method.</Paragraph>
      <Paragraph position="2"> Firstly, the user often needs to press a key numerous times to obtain a single target character. Secondly, there is an ambiguity in the user action when two characters assigned to the same button are to be entered one after another ('aa' requires the entry of '11' that can also be 'b'). This situation requires the use of an escape.</Paragraph>
      <Paragraph position="3"> A second possible method is to press a rst button to select it, and then enter the number i to select the ith character which is assigned to the rst button.</Paragraph>
      <Paragraph position="4"> For example, on many mobile phones, `o' is obtained by pressing the no. 6 key and then the no. 3 key, since `o' is the third letter on the no. 6key. However, if the number of letters on each key is greater than the number of keys, entry of the higher i values is implausiblydicult. With theTouchMeKey4system, for example, a system for the easy entry of fth and sixth characters, etc., is not possible.</Paragraph>
      <Paragraph position="5"> In short, the free entry of text turns out to be too dicult with a four-button device unless we adopt  As with any system where a predictive method is applied, the weak pointofTouchMeKey4 is the processing of unknown words which do not appear in the dictionary. Therefore, it is important that the Base dictionary contains a richvocabulary. When, however, an unknown word occurs, it may still be entered character bycharacter by using the methods described inx3.1, or the system may be connected with a larger dictionary via a network.</Paragraph>
      <Paragraph position="6">  (no. wrds) No. di. wrds - 868 761 in test doc.</Paragraph>
      <Paragraph position="7"> prediction. This is so even for the case of English, the written form of which has relatively few characters, andiseven more so for languages with large numbers of characters such as Chinese, Japanese, or Thai (78 characters). We are thus obliged to use prediction.</Paragraph>
    </Section>
    <Section position="2" start_page="2" end_page="2" type="sub_section">
      <SectionTitle>
3.2 Predictive Text entry
</SectionTitle>
      <Paragraph position="0"> Generally, there are twoways to predict candidates.</Paragraph>
      <Paragraph position="1"> The rst is the single-tap method. The earliest appearance of this idea was at the beginning of the 80's in Japan, in discussions of processing systems for Japanese text(Co.Ltd., 1982);; more recent work has been concerned with mobile phones (James and Reischel, 2001)(Tanaka-Ishii et al., 2000).</Paragraph>
      <Paragraph position="2"> The second wayisprediction by prefix. Given a user input, the system searches for words with the corresponding prefix.</Paragraph>
      <Paragraph position="3"> This method of collecting candidates to be oered to the user has been particularly successful in the entry of Chinesetext. The method has alsobeen applied to certain text-entry systems in the man-machine interface domain, too (Masui, 1999).</Paragraph>
      <Paragraph position="4"> As the description of x2 indicates, the combination of the two methods is adopted in our TouchMeKey4 system. It thus needs to process many candidates for a single user entry. The mechanism of estimating levels of relevance for the words is explained in the next section.</Paragraph>
    </Section>
  </Section>
  <Section position="6" start_page="2" end_page="2" type="metho">
    <SectionTitle>
4 Applying an Adaptive Language
</SectionTitle>
    <Paragraph position="0"> Model in Candidate Estimation As was summarized in x2, the PPM (prediction by partial match) framework is used by TouchMeKey4 to estimate the relevance of candidates. Its characteristic is that the word distribution is adapted to the style of the user's corpus.</Paragraph>
    <Paragraph position="1"> PPM was originally proposed as an adaptive language model for use in improving the compression rates of arithmetic coding. The estimation of probabilities by PPM thus guarantees a lowering of the entropy of the language model. PPM has successfully been adapted to the user-interface domain in certain previous works(Tanaka-Ishii et al., 2001)(Ward et al., 2000).</Paragraph>
    <Paragraph position="2"> Broadly, PPM interpolates the n-gram counts in the user corpus and the statistics in the base dictionary. The following formula is used to estimate a probability for the ith word w</Paragraph>
    <Paragraph position="4"> Here, k, the order, indicates the number of words be-</Paragraph>
    <Paragraph position="6"> that are used in the calculation of P</Paragraph>
    <Paragraph position="8"> probability that is obtained from the base dictionary.</Paragraph>
    <Paragraph position="9"> For other k, P</Paragraph>
    <Paragraph position="11"> ) is calculated from statistics obtained from User corpus. Finally, u</Paragraph>
    <Paragraph position="13"> ). Of the manystudies of u k (Teahan, 2000), we have chosen PPM-A(Bell et al., 1990), the simplest, because our preliminary experiments showed no signicant dierence in performance among the methods we tried.</Paragraph>
    <Paragraph position="14"> We decided to utilize this PPM framework because the context is the most suitable item of information for the elimination of irrelevant candidates. Small machines are in a personal context, and oce and household machines are used in particular contexts. With this method, the language model is adaptable on the y. This is achieved by simply accumulating the user's newly entered text at the end of the user corpus.</Paragraph>
    <Paragraph position="15"> In this paper, the Base dictionary contains the uni-gram probabilities obtained from Wall Street Journal as was explained in x2. We prepared various User corpora,: three in English, three in Japanese and two in Thai. Of these, the characteristics of two of the English User corpora that are used in x6aregiven in</Paragraph>
  </Section>
  <Section position="7" start_page="2" end_page="3" type="metho">
    <SectionTitle>
5 Key Assignment
</SectionTitle>
    <Paragraph position="0"> The assignment of characters to the respective buttons is one determinant of the eciency of text entry. For example, if all characters from `a' to `w' are assigned to the rst key and `x', `y', and `z' are respectively assigned to the second, third and fourth keys, the performance in word prediction will clearly be bad. The problem of key assignment remains even when we have eliminated such extreme possibilities,  because there are many plausible assignments. We thus need to be able to measure the performance of a key assignment.</Paragraph>
    <Paragraph position="1"> One way to measure this is to experimentally decide it by automatically entering some documents (as will be described in the x6 later in this paper). However, the result of such a test is dependent on the test document which is used. Lower-level settings, such as key assignments, should, as much as is possible, be for general-purpose use.</Paragraph>
    <Paragraph position="2"> Having key sequences as C and the target word as W, the task of the system is to estimate a better W from C. Information theory provides us with a tool for estimating the uncertaintyofthis task: the average conditional entropy. The denition of this</Paragraph>
    <Paragraph position="4"> quence c and P(W = wjC = c) is the conditional probability of words for the given c. When the estimation of W is less certain, H(WjC) has a larger value. The lower the entropy, the less uncertain the estimation of the word. Therefore, the conditional entropy is suitable as a method for the evaluation of key assignments.</Paragraph>
    <Paragraph position="5"> One other factor that we need to consider at this point is the order of the alphabet. English has an alphabet order that even children know. If this order is neglected and the letters `ajxgukh' are assigned to a given key, the interface will become dicult for the beginners, although it might be the most ecientfor a well-trained user. Therefore, the key assignments had better reect such linguistic tradition.</Paragraph>
    <Paragraph position="6"> We took this into consideration in generating some possible key assignments. Table 2 is a list of the assignments and their values of conditional entropyas calculated on the basis of one year of issues of WSJ.</Paragraph>
    <Paragraph position="7"> The rst column shows the total number of keys (below denoted by N</Paragraph>
    <Paragraph position="9"> ). We here consider the situations where there are ve, three, and two, as well as four, buttons. The second column gives a label for each of the key assignments. In the third column, the characters to be assigned to the respective buttons are grouped in parentheses. For example, 4-A indicates an assignment to four keys with 'abcdef' assigned to the rst button, 'ghijkl' to the second button, 'mnopqrs' to the third, and 'tuvwxyz' and other ASCII symbols to the fourth. The capital letters are assigned to the same keys as the corresponding small  . Note that 4-A corresponds to the TouchMeKey4 assignment whichwesaw in Figure 1-1. The groupings with the label C are more random than those with other two.</Paragraph>
    <Paragraph position="10"> In general, entropyvalues fall as the number ofkeys increases. This is a readily comprehensible result;; a larger number of keys eases the task of estimation, thus making it less uncertain. When we compare the values for assignments to the same numbers of keys, we see that the entropyvalues dier considerably. For example, the entropy of 5-B indicates more uncertainty than the other 5-x assignments. The entropy value is the same as for 4-A, although the number of keys in use is dierent (this is comprehensible when we look at the similar character groupings of 4-A and 5-B).</Paragraph>
    <Paragraph position="11"> In this paper, we evaluate the use of key assignments with the label A on TouchMeKey4, since they havelower entropyvalues than the other settings.</Paragraph>
  </Section>
class="xml-element"></Paper>