<?xml version="1.0" standalone="yes"?>
<Paper uid="C02-1048">
  <Title>Answering it with charts -- Dialogue in natural language and charts --</Title>
  <Section position="3" start_page="0" end_page="0" type="metho">
    <SectionTitle>
2 Context sensibility of chart
</SectionTitle>
    <Paragraph position="0"> appropriateness Let us suppose an analyst, planning sales of her company's products, gets interested in its sales in a particular district. So she requests the following from a system in front of her:  (1) Show me the sales in the Shikoku district for '93 and '94.</Paragraph>
    <Paragraph position="1"> The system answers this request by drawing an appropriate chart. She continues and utters a fragment of a request: (2) By prefecture.</Paragraph>
    <Paragraph position="3"/>
    <Paragraph position="5"> The system understands this fragment and draws a new chart. This dialogue might be followed by an utterance like: (3) Through '98.</Paragraph>
    <Paragraph position="6"> The correct chart made in the response to utterance (1) is like that shown in Figure 1-(1). Following this chart, the response to utterance (2) should be made by the stacked column chart, as shown in Figure 1-(2), in which each column is subdivided in order to show the sales by prefecture. Here, these two consecutive utterances have a combined meaning similar to: (2') Show me the sales in each prefecture of the Shikoku district for '93 and '94.</Paragraph>
    <Paragraph position="7"> It is interesting that when utterance (2') is given without a specific context, the response should be made by the grouped column chart, as shown in Figure 1-(2'), rather than as in Figure 1-(2). The preference for Figure 1-(2) as the response to utterance (2) may come from the perspective represented in utterance (1) that she wants to look at the sales of the district in total or the similarity in shape between Figure 1-(2) and 1-(1). In any case, it is important that an appropriate chart form depends on what utterance or series of utterances was used to express a request and what chart has been drawn previously.</Paragraph>
    <Paragraph position="8"> This context sensibility of chart appropriateness oc-</Paragraph>
    <Paragraph position="10"/>
    <Paragraph position="12"> curs constantly. For utterance (3) in the previous dialogue, the chart shown in Figure 1-(3) is preferred when Figure 1-(2) is used to answer the previous request. The chart of Figure 1-(3') is preferred, however, as the response to utterance (3'), which combines requests (1), (2) and (3) into one; (3') Show me the sales in each prefecture of Shikoku district from '93 through '98.</Paragraph>
    <Paragraph position="13"> The chart form is not the only dimension sensitive to dialogue context. Consider the following example. null  (4) Show me the sales in Shikoku and Chugoku for '93.</Paragraph>
    <Paragraph position="14"> (5) Add the one for '94.</Paragraph>
    <Paragraph position="15"> (5') Show me the sales in Shikoku and Chugoku for '93 and '94.</Paragraph>
    <Paragraph position="16">  The preferred response to utterance (5) must be in the chart shown in Figure 2-(2) when preceding utterance (4) was answered by the chart shown in Figure 2-(1), while the chart of Figure 2-(2') would be used when the request is just utterance (5'). It is clear that the decision on axis assignment is also context sensitive.</Paragraph>
    <Paragraph position="17"> In conventional ellipsis handling (Hendrix et al., 1978; Carbonell and Hayes, 1983), the interpretation of an utterance fragment, such as utterance (2) following utterance (1), is the same as the interpretation of utterance (2'). When a response based on this interpretation is made, the data plotted on the chart may be correct, but the chart form and style cannot be. This implies that something extra is needed for handling dialogue in charts, and it is insufficient just to combine two mechanisms for non-interactive automatic chart design and natural language dialogue understanding.</Paragraph>
    <Paragraph position="18"> 3 Handling dialogue in natural language and charts This section proposes a methodology for handling dialogue in natural language and charts. First, a logical form that represents the interpretation of utterances is proposed. Then, how to represent the perspectives from which the user wants to look at the data and how to relate them to chart realization are described. Last, a way of handling utterance fragments is discussed.</Paragraph>
    <Section position="1" start_page="0" end_page="0" type="sub_section">
      <SectionTitle>
3.1 A logical form
</SectionTitle>
      <Paragraph position="0"> A logical form that describes the content of a given utterance must be able to represent what region of data the user is interested in and what type of analysis she wants to conduct in order to obtain the data to be plotted. Moreover, it must convey the intention of the user, that is, what information she wants to obtain through that chart. The logical form we propose, which satisfies those requirements, takes the form, a73 Description, a74a75a74a75a74a75a76 Actiona77 , where Action specifies the main speech act in a given utterance, and Descriptions describe constraints or conditions that the objects related to the action should satisfy. Action can be a request to display charts or a request for information conveyed through charts. In this paper, however, it only covers requests to display charts, which takes the form a78a80a79a82a81a82a83a85a84a87a86a89a88a85a90a87a91a93a92a95a94 ListofVars, ListofAspectsa96 , where ListofVars is the list of variables plotted on the chart. ListofAspects is the list of aspects of the data the user is focusing on and represents the perspectives from which she wants to look at the data.</Paragraph>
      <Paragraph position="1"> Descriptions describe constraints or conditions that the objects related to the action should satisfy, which has the form, a73 Quantifier, Var/Class, Restrictiona77 , where Quantifier is a generalized quantifier, Var is the variable of quantification, and the quantification ranges over the objects each of which is a member of Class and satisfies Restriction. That is, this logical form is a flattened version of Woods' MRL (Woods, 1978), and as in Woods' MRL, Class can be a function. Moreover, Classes, each of which each variable and object is associated with, are hierarchically organized and represent not only the domain an object is classified into, but also its granularity. An object that belongs to the area domain, for example, belongs to one of classes: district, prefecture, or city, according to its granularity. The subsumption relation is defined between objects that belong to classes with different granularities and the same domain. In Restriction, implicit coercion between granularities is allowed and aggregation such as summation is represented implicitly using this mechanism. 1 For example, utterance (2') is interpreted into the logical form:</Paragraph>
      <Paragraph position="3"> The first description states that variable a149 ranges over two objects of year class, 1993 and 1994. The second description states variable a150 ranges over objects with prefecture granularity that are subsumed by Shikoku, which is itself an object with district granularity. In this case, the equality in the restriction coerces into the subsumption relation. In the third description, a86a82a91a87a90a87a79a80a86a151a94a109a149a153a152a113a150a154a96 is a function from time and area to sales amounts. The perspective is discussed in the next section.</Paragraph>
    </Section>
    <Section position="2" start_page="0" end_page="0" type="sub_section">
      <SectionTitle>
3.2 Perspectives and chart realization
</SectionTitle>
      <Paragraph position="0"> The second argument of the a78a80a79a82a81a82a83a85a84a87a86a89a88a85a90a87a91a93a92 action specifies the user's perspective, which is the list of aspects of the data she is focusing on. The following basic aspects are covered for the present. Suppose Var1 is an element of the first argument of the action, that is, a variable to be plotted. The a155a82a156a108a157 a88a80a158a87a92a95a94 Var2a96 specifies that the user is interested in comparison of the values of Var1 by each possible Var2 instance. In other words, she is focusing on the difference in the values of Var1 which came from the difference of Var2 instances. The  she is interested in the proportion of each value of Var1 to the total in all Var2 instances. The</Paragraph>
      <Paragraph position="2"> in the change of value of Var1 with the change of Var2 instances. The a155a82a156a108a157 a88a80a158a87a92a95a94 Var2a96 is, in a sense, a default since the quantitative chart, in principle, shows variation of the values of a dependent variable by the difference of the values of independent variables.</Paragraph>
      <Paragraph position="3">  in (Matsushita et al., 2000).</Paragraph>
      <Paragraph position="4"> From now on, our discussion is restricted to the two-dimensional chart forms for up to two independent variables and one dependent variable. These forms of charts are widely used and still have plenty of variety. In chart realization under this restriction, one of the independent variables, which are the arguments of the function in the logical form, is assigned to the horizontal axis and the other to the hidden or depth axis. For example, variable a149 is assigned to the horizontal axis and variable a150 is assigned to the depth axis in the realization of the logical form of the previous example into Figure 1(2'). One of the realization parameters is this axis assignment of variables.</Paragraph>
      <Paragraph position="5"> Let us think about how the perspective guides chart realization. In two-dimensional charts, changes of values can be displayed along the horizontal axis using, say, the line chart, while total and proportion is easy to show on a depth axis by using, say, the stacked chart. Comparison can be exhibited on either axis. Therefore, a155a82a156a108a157 a88a80a158a87a92 and</Paragraph>
      <Paragraph position="7"> a163 are possible for the variable assigned to the horizontal axis, and a155a82a156a108a157 a88a80a158a87a92 , a159 a156 a159 a91a87a90a80a160a89a161 , and</Paragraph>
      <Paragraph position="9"> a156 are possible for the variable assigned to the depth axis. Next, let us think about what combinations of aspects are possible and sufficient for the perspective. Since we cannot compare the difference nor see the changes regarding a variable that has only one instance in the range specified by the restriction, the aspect for such a variable, which we call a uniquely instantiated variable, is meaningless. Excluding that variable, each remaining variable takes only one aspect. Their combination should result in those aspects being properly assigned to the axes. Here, by subcategorizing  to the horizontal and depth axis respectively, you can assign an axis to the variables only by assigning one aspect to each. Thus, when two independent variables, a149 and a150 , are not uniquely instantiated, one of them, say a149 , takes either a163a104a166a80a156a108a157 a88a80a158a87a92a95a94a109a149a85a96</Paragraph>
      <Paragraph position="11"> this case, variable a149 is assigned to the horizontal axis and variable a150 to the depth axis. Both variables are not allowed to take aspects assigned to the same axis. The combination of a167</Paragraph>
      <Paragraph position="13"> a91a87a90a80a160a89a161a153a94a109a150a154a96 is forbidden, for example. Figure 3 summarizes the relationship between combinations of aspects and realized chart forms.</Paragraph>
      <Paragraph position="14"> Now, chart realization is reduced to aspect assignment. In other words, you can realize an appropriate chart from a given user utterance by obtaining its logical form, and, if some of the aspects are implicit in the utterance, by completing the perspective using the characteristics of the data concerned, the current context and so on.2 Obtaining and completing the perspective, which is vital for chart realization, is carried out as follows.3 a168 A portion of the perspective is explicit in the utterance. For example, it is obvious that the user is focusing on the change over time when she says that she wants to see the change in sales. Such perspectives are acquired through the interpretation of her utterance.</Paragraph>
      <Paragraph position="15"> a168 Characteristics of statistics returning a dependent variable sometimes suggest the perspective. For example, share and profitability suggest her interest in the proportion to the total.</Paragraph>
      <Paragraph position="16"> Such characteristics are used for completing the perspective.</Paragraph>
      <Paragraph position="17"> a168 The class of independent variables also suggests the perspective. A class whose instances constitute an interval scale, such as a time series, suggests changes over it (i.e. a155a89a163 a91a108a161a104a164a80a79a93a165a85a84 a159 a163 ) for its aspect to be focused on when the variable ranges over a lot of instances. Even when it has a few instances, comparisons by it (i.e. a163a104a166a80a156a108a157 a88a80a158a87a92 ) are preferred. This criterion for selecting between  cussed here. For example, while the independent variable is always assigned to the vertical axis in our discussion, it can be assigned to the horizontal axis. The rank of instances on an axis, the scales of axes, and visual prompts such as labels and arrows are also dimensions which should be considered (Mittal, 1998; Fasciano and Lapalme, 1996). Although discussion of those dimesions exceeds the scope of this paper, we believe that a natural extension of perspective would cover them.</Paragraph>
    </Section>
    <Section position="3" start_page="0" end_page="0" type="sub_section">
      <SectionTitle>
3.3 Utterance fragments and chart
appropriateness
</SectionTitle>
      <Paragraph position="0"> Utterance fragments in a specific context should be interpreted not as a logical form, but as a request to revise the logical form given as the context. In order to correctly handle chart appropriateness, which is sensitive to the context, the perspective in the logical form should be revised appropriately according to that request. First, utterance fragments are classified into the following categories according to what part of the logical form should be revised as a result of those fragments. Examples are shown, which are assumed to have been uttered after the utterance, &amp;quot;Show me the sales in Shikoku and Chugoku from  '93 through '95.&amp;quot; Domain alteration: The Restriction part of an independent variable is revised. Examples, &amp;quot;Just Shikoku.&amp;quot;, &amp;quot;Through '98.&amp;quot; Granularity alteration: The Class part of an independent variable is revised to one with a different granularity. Examples, &amp;quot;By quarter.&amp;quot;, &amp;quot;By prefecture.&amp;quot; Statistic alteration: The function returning the dependent variable, which locates its Class part, is revised. Examples, &amp;quot;Show me the number of the dealers.&amp;quot; Perspective alteration: The perspective is revised.</Paragraph>
      <Paragraph position="1"> Examples, &amp;quot;Show me the change.&amp;quot;, &amp;quot;How  about the total?&amp;quot; The type of content words, clue words, and specific phrases contained are exploited for interpreting utterance fragments and for classifying them into one of the above categories. Using those, we can identify what part of the logical form should be changed and how. In addition to the revisions identified, appropriate revision of the perspective is needed for correct chart realization. Revisions of perspective are summarized as follows.4 a168 As a result of domain alteration on variable X, if the number of instances of X turns into more than one and the current perspective includes no aspect relating to X, that is, X is a uniquely instantiated variable, check a155a89a163 a91a108a161a104a164a80a79a93a165a85a84  and add the first possible one to the perspective. On the other hand, if X turns into a uniquely instantiated variable, delete the aspect related to X from the perspective.</Paragraph>
      <Paragraph position="2"> a168 As a result of granularity alteration on variable X, if the number of instances of X turns into more than one and the current perspective includes no aspect related to X, check a155a89a163 a91a108a161a104a164a80a79a93a165a85a84  and add the first possible one to the perspective. Here, a159 a156 a159 a91a87a90a80a160a89a161a153a94 Xa96 is possible only when the statistics concerned use summation for aggregation. On the other hand, if X turns into 4Revisions of perspective for statistic alterations and perspective alterations are omitted, because the space is limited and our concern is to trace the changes of the user's perspective especially when she does not mention them explicitly. a uniquely instantiated variable, delete the aspect related to X from the perspective.</Paragraph>
    </Section>
  </Section>
  <Section position="4" start_page="0" end_page="0" type="metho">
    <SectionTitle>
4 Examples
</SectionTitle>
    <Paragraph position="0"> This section demonstrates how our proposal addresses the problems raised. First, let us consider series of utterance (1), (2) and (3). The interpretation of utterance (1) is  As for the perspective, variable a149 , which represents time series and ranges over only two instances, obtains not a155a89a163 a91a108a161a104a164a80a79a93a165a85a84 a159 a163 a94a109a149a154a96 , but a163a104a166a80a156a108a157 a88a80a158a87a92a95a94a109a149a154a96 . Variable a150 , which ranges over areas, is uniquely instantiated and no aspect is given to it. The chart realized from this logical form is the column chart of Figure 1-(1). The chart form is determined from the perspective by referring to Figure 3. When utterance (2) is given in this context, it is interpreted as a granularity alteration on variable a150 , and  tion, a159 a156 a159 a91a87a90a80a160a89a161a153a94a109a150a154a96 is added to the perspective, since no aspect related to a150 was in it and sales is a statistic for which summation is used for aggregation. The perspective ends up with a247 a163a104a166a80a156a108a157 a88a80a158a87a92a95a94a109a149a154a96a154a152  a91a87a90a80a160a89a161a153a94a109a150a154a96a89a248 , and the new chart is the stacked column chart of Figure 1-(2). Utterance (3) in this context is interpreted as a domain alteration on variable a149 . Since the perspective remains the same as before, while the restriction of variable a149 is revised according to the utterance, the chart obtained is of the same form, which is shown in Figure 1-(3).</Paragraph>
    <Paragraph position="1"> On the other hand, for utterance (2'), as neither its expression nor its statistic implies a specific aspect, the perspective is determined according to the characteristics of the independent variables.</Paragraph>
    <Paragraph position="2"> First, variable a149 representing time series obtains a163a104a166a80a156a108a157 a88a80a158a87a92a95a94a109a149a154a96 , and then variable a150 ranging over areas is given a167  a88a80a158a87a92a95a94a109a150a154a96a108a248 , and the chart realized is the grouped column chart shown in Figure 1-(2'). For utterance (3'), almost the same criteria are applied, but the aspect given to variable a149</Paragraph>
    <Paragraph position="4"> a94a109a149a85a96 as it has many instances. As a result, the grouped line chart of Figure 1-(3') is realized. null Let us move to series of utterance (4) and (5). For utterance (4), since variable a149 representing time series is uniquely instantiated, no aspect is given. Then variable a150 ranging over districts can obtain a163a104a166a80a156a108a157 a88a80a158a87a92a95a94a109a150a154a96 . By utterance (5) following it, domain alteration on a149 is specified, and a149 obtains the possible aspect a167 a166a80a156a108a157 a88a80a158a87a92a95a94a109a149a154a96 , since a163a104a166a80a156a108a157 a88a80a158a87a92a95a94a109a150a85a96 already exists. On the other hand, for utterance (5'), variable a149 representing time series priors variable a150 , obtaining a163a104a166a80a156a108a157 a88a80a158a87a92a95a94a109a149a154a96 . As a result, Figure2-(2), the response to utterance (5), and Figure 2-(2') , the response to utterance (5'), are different in their axis assignments.</Paragraph>
  </Section>
class="xml-element"></Paper>