<?xml version="1.0" standalone="yes"?>
<Paper uid="E06-2030">
  <Title>Developments in Affect Detection in E-drama</Title>
  <Section position="3" start_page="203" end_page="204" type="metho">
    <SectionTitle>
2 A Preliminary Approach
</SectionTitle>
    <Paragraph position="0"> Various characterizations of emotion are used in emotion theories. The OCC model uses emotion labels and intensity, while Watson and Tellegen (1985) use positive and negative affects as the major dimensions. Currently, we use an evaluation dimension (positive and negative), affect labels and intensity. Affect labels with intensity are used when strong text clues signalling affect are detected, while the evaluation dimension with intensity is used when only weak text clues are detected.</Paragraph>
    <Section position="1" start_page="203" end_page="203" type="sub_section">
      <SectionTitle>
2.1 Pre-processing Modules
</SectionTitle>
      <Paragraph position="0"> The language in the speeches created in e-drama sessions, especially by excited children, severely challenges existing language-analysis tools if accurate semantic information is sought. The language includes misspellings, ungrammaticality, abbreviations (such as in texting), slang, use of upper case and special punctuation (such as repeated exclamation marks) for affective emphasis, repetition of letters or words for emphasis, and open-ended onomatopoeic elements such as &amp;quot;grrrr&amp;quot;. The genre is similar to Internet chat. To deal with the misspellings, abbreviations and onomatopoeia, several pre-processing modules are used before the detection of affect starts using pattern matching, syntactic processing by means of the Rasp parser (Briscoe &amp; Carroll, 2002), and subsequent semantic processing.</Paragraph>
      <Paragraph position="1"> A lookup table has been used to deal with abbreviations e.g. 'im (I am)', 'c u (see you)' and 'l8r (later)'. It includes abbreviations used in Internet chat rooms and others found in an anlysis of previous edrama sessions. We handle ambiguity (e.g.,&amp;quot;2&amp;quot; (to, too, two) in &amp;quot;I'm 2 hungry 2 walk&amp;quot;) by considering the POS tags of immediately surrounding words. Such simple processing inevitably leads to errors, but in evaluations using examples in a corpus of 21695 words derived from previous transcripts we have obtained 85.7% accuracy, which is currently adequate.</Paragraph>
      <Paragraph position="2"> The iconic use of word length (corresponding roughly to imagined sound length) as found both in ordinary words with repeated letters (e.g.</Paragraph>
      <Paragraph position="3"> 'seeeee') and in onomatopoeia and interjections, (e.g. 'wheee', 'grr', 'grrrrrr', 'agh', 'aaaggghhh') normally implies strong affective states. We have a small dictionary containing base forms of some special words (e.g. 'grr') and some ordinary words that often have letters repeated in e-drama.</Paragraph>
      <Paragraph position="4"> Then the Metaphone spelling-correction algorithm, which is based on pronunciation, works with the dictionary to locate the base forms of words with letter repetitions.</Paragraph>
      <Paragraph position="5"> Finally, the Levenshtein distance algorithm with a contemporary English dictionary deals with misspelling.</Paragraph>
    </Section>
    <Section position="2" start_page="203" end_page="204" type="sub_section">
      <SectionTitle>
2.2 Affect Detection
</SectionTitle>
      <Paragraph position="0"> In the first stage after the pre-processing, our affect detection is based on textual pattern-matching rules that look for simple grammatical patterns or phrasal templates. Thus keywords, phrases and partial sentence structures are extracted. The Jess rule-based Java framework is used to implement the pattern/template-matching rules. This method has the robustness to deal with ungrammatical and fragmented sentences and varied positioning of sought-after phraseology, but lacks other types of generality and can be fooled by suitable syntactic embedding. For example, if the input is &amp;quot;I doubt she's really angry&amp;quot;, rules looking for anger in a simple way will output incorrect results.</Paragraph>
      <Paragraph position="1"> The transcripts analysed to inspire our initial knowledge base and pattern-matching rules had independently been produced earlier from edrama improvisations based on a school bullying scenario. We have also worked on another, distinctly different scenario concerning a serious disease, based on a TV programme produced by Maverick Television Ltd. The rule sets created for one scenario have a useful degree of applicability to another, although some changes in the specific  knowledge database will be needed.</Paragraph>
      <Paragraph position="2"> As a simple example of our pattern-matching, when the bully character says &amp;quot;Lisa, you Pizza Face! You smell&amp;quot;, the module detects that he is insulting Lisa. Patterns such as 'you smell' have been used for rule implementation. The rules work out the character's emotions, evaluation dimension (negative or positive), politeness (rude or polite) and what response EmEliza might make. Although the patterns detected are based on English, we would expect that some of the rules would require little modification to apply to other languages.</Paragraph>
      <Paragraph position="3"> Multiple exclamation marks and capitalisation of whole words are often used for emphasis in edrama. If exclamation marks or capitalisation are detected, then emotion intensity is deemed to be comparatively high (and emotion is suggested even without other clues).</Paragraph>
      <Paragraph position="4"> A reasonably good indicator that an inner state is being described is the use of 'I' (see also Craggs and Wood (2004)), especially in combination with the present or future tense. In the school-bullying scenario, when 'I' is followed by a future-tense verb, a threat is normally being expressed; and the utterance is often the shortened version of an implied conditional, e.g., &amp;quot;I'll scream [if you stay here].&amp;quot; When 'I' is followed by a present-tense verb, other emotional states tend to be expressed, as in &amp;quot;I want my mum&amp;quot; and &amp;quot;I hate you&amp;quot;. Another useful signal is the imperative mood, especially when used without softeners such as 'please': strong emotions and/or rude attitudes are often being expressed. There are common imperative phrases we deal with explicitly, such as &amp;quot;shut up&amp;quot; and &amp;quot;mind your own business&amp;quot;. But, to go beyond the limitations of the pattern matching we have done, we have also used the Rasp parser and semantic information in the form of the semantic profiles for the 1,000 most frequently used English words (Heise, 1965).</Paragraph>
      <Paragraph position="5"> Although Rasp recognizes many simple imperatives directly, it can parse some imperatives as declaratives or questions. Therefore, further analysis is applied to Rasp's syntactic output.</Paragraph>
      <Paragraph position="6"> For example, if the subject of an input sentence is 'you' followed by certain special verbs or verb phrases (e.g. 'shut', 'calm', 'get lost', 'go away', etc), and Rasp parses a declarative, then it will be changed to imperative. If the softener 'please' is followed by a base forms of the verb, the inputs are also deemed to be imperatives. If a singular proper noun or 'you' is followed by a base form of the verb, the sentence is deemed to be imperative (e.g. &amp;quot;Dave bring me the menu&amp;quot;). When 'you' or a singular proper noun is followed by a verb whose base form equals its past tense form, ambiguity arises (e.g. &amp;quot;Lisa hit me&amp;quot;). For one special case of this, if the direct object is 'me', we exploit the evaluation value of the verb from Heise's (1965) semantic profiles. Heise lists values of evaluation (goodness), activation, potency, distance from neutrality, etc. for each word covered. If the evaluation value for the verb is negative, then the sentence is probably not imperative but a declarative expressing a complaint (e.g &amp;quot;Mayid hurt me&amp;quot;). If it has a positive value, then other factors suggesting imperative are checked in this sentence, such as exclamation marks and capitalizations. Previous conversation is checked to see if there is any recent question sentence toward the speaker. If so, then the sentence is taken to be declarative.</Paragraph>
      <Paragraph position="7"> There is another type of sentence: 'don't you + (base form of verb)', which is often a negative version of an imperative with a 'you' subject (e.g.</Paragraph>
      <Paragraph position="8"> &amp;quot;Don't you call me a dog&amp;quot;). Normally Rasp regards such strings as questions. Further analysis has also been implemented for such sentence structure, which implies negative affective state, to change the sentence type to imperative.</Paragraph>
      <Paragraph position="9"> Aside from imperatives, we have also implemented simple types of semantic extraction of affect using affect dictionaries and WordNet.</Paragraph>
    </Section>
  </Section>
  <Section position="4" start_page="204" end_page="205" type="metho">
    <SectionTitle>
3 Metaphorical Expression of Affect
</SectionTitle>
    <Paragraph position="0"> The explicit metaphorical description of emotional states is common and has been extensively studied (Fussell &amp; Moss, 1998). Examples are &amp;quot;He nearly exploded&amp;quot;, and &amp;quot;Joy ran through me.&amp;quot; Also, affect is often conveyed implicitly via metaphor, as in &amp;quot;His room is a cess-pit&amp;quot;, where affect associated with a source item (cess-pit) is carried over to the corresponding target item.</Paragraph>
    <Paragraph position="1"> Physical size is often metaphorically used to emphasize evaluations, as in &amp;quot;you are a big bully&amp;quot;, &amp;quot;you're a big idiot&amp;quot;, and &amp;quot;you're just a little bully&amp;quot;, although the bigness may be literal as well. &amp;quot;Big bully&amp;quot; expresses strong disapproval (Sharoff, 2005) and &amp;quot;little bully&amp;quot; can express contempt, although &amp;quot;little&amp;quot; can also convey sympathy. Such examples are not only practically important but also theoretically challenging.</Paragraph>
    <Paragraph position="2"> We have also encountered quite creative use of metaphor in e-drama. For example, in a school-bullying improvisation that occurred, Mayid had already insulted Lisa by calling her a 'pizza', developing a previous 'pizza-face' insult. Mayid then said &amp;quot;I'll knock your topping off, Lisa&amp;quot; - a theoretically intriguing spontane- null ous creative elaboration of the 'pizza' metaphor.</Paragraph>
    <Paragraph position="3"> Our developing approach to metaphor handling in the affect detection module is partly to look for stock metaphorical phraseology and straight-forward variants of it, and partly to use a simple version of the more open-ended, reasoning-based techniques taken from the ATT-Meta project (Barnden et al., 2002; 2003; 2004). ATT-Meta includes a general-purpose reasoning engine, and can potentially be used to reason about emotion in relation to other factors in a situation. In turn, the realities of metaphor usage in e-drama sessions are contributing to our basic research on metaphor processing.</Paragraph>
  </Section>
class="xml-element"></Paper>