<?xml version="1.0" standalone="yes"?>
<Paper uid="W97-0714">
  <Title>ized with Macromed\]a D~rector CommumcaUon between CLOS and Macromecha Director Is medmted by Apple Events</Title>
  <Section position="4" start_page="89" end_page="91" type="metho">
    <SectionTitle>
2 System overview
</SectionTitle>
    <Paragraph position="0"> SlmSum currently runs on Macintoshes with .System 7.5, a CD-drlve, a 17&amp;quot; momtor and some addmonal RAM as Is usual for multtme&amp;a apphcatlons It is Implemented as. an obJect-oriented blackboard system m CLOS and Macromedla Director (see figs 1 and 2) CogmUve strategies are represented by object-oriented agents grouped around thetr respective blackboards The agents are eqmpped wRh specmhzed knowledge, e g an indicator phrase lexicon or a basic representation of SGML codes They process text structure m an SGML-hke coding and text meaning m a proposmonal representation The fact knowledge referenced m texts Is defined m document-specific ontologles On the screen the agents appear as ammated beasts The CLOS obJects simulate the cogmtlve strategies They send out Apple Events to make &amp;quot;their&amp;quot; ammals on the stage perform according to the stmulatton An access system accommodates user mteract~on..m a movte-hke style the user chooses a workzng, sequence and can interrupt at any time to get further explanaUons about what the cogmtlve agents do, how they are motivated empmcaUy and how they are Imple- null Figure 2 gives a screenshot of the. SlmSum multlmedra interface, presenting the relevance assessment agents at work The agent relevant-texthmt (a ladybird) is putting its can&amp;date statements on the relevance blackboard, whale the relevance agents hold and relevant-umt are s~tung on the bench, together with the suspended control agents explore for document explorataon and understart(brig (the bee) and construct for target text producuon (the spider) Below, can we see the document blackboard with the representaUon of the source text, showing ~ts meamng panel, the scheme representation stating the document orgamzaUon, and the theme representaUon stonng the theme, ~ e the top of the macrostructure as far as known to the summarizer At the bottom, a comment explains what is. currently happemng on the screen The central system components have been derived from observation</Paragraph>
    <Section position="1" start_page="90" end_page="91" type="sub_section">
      <SectionTitle>
* Cooperating agents
Experts use recurnng goal-oriented proce-
</SectionTitle>
      <Paragraph position="0"> dures, corresponding to the strategies sketched by van Dijk and Kmtsch (1983) These procedures or strategies were operat~onahzed into mtelhgent agents of the computerized system Agents consist of a script that defines how they deal with tbelr task, they have a general cornmumcauon component that allows them to exchange messages with other agents and to access global knowledge sources, they may possess private task-oriented knowledge, and they are eqtupped with task-onented data wews for input and output Control agents (responsible agents for a blackboard - see below) are m ad&amp;uon assigned a little scheduler They actlvate thetr subordinates by d~rect message passing Data exchange between agents takes place v~a the blackboards The agents keep to the commumcat~on rules Strategies / agents cooperate in concrete tasks such ass dectdmg about relevance or semng up a target summary statement Agents may use products of other agents, but since they have hmlted tasks, they have no soplust~cated commumcataon behavtour such as bargaining or &amp;scussing null Hgure 2 A screenshot of the SlmSurn user interface relevance assessment agents 0adyblrds) are busy wlule exploraUon (done by bees) and target text producUon (by spiders) are suspended  s ' Blackboards Agents need commumcatlon areas, as a medium of cooperation Functionally speaking, these are blackboards (Selfndge, 1959, Carver &amp; Lesser, 1994, Engelmore &amp; Morgan, 1988) StmSum blackboards are dedicated They are used for reception, storage of the input text representation, relevance assessment, target summary construction and so on Central is the document blackboard that stores and organizes all knowledge acqmred from the source document (of fig 2) Since m the case of professional summarizing cogmtlve processing is modular, the agents work m task-specific groups using a dechcated blackboard For instance, the relevance assessment agents use the relevance blackboard to put the relevance judgement together Every blackboard has a control specialist It orgamzes the work of the group, sums up what they have acineved, executes.</Paragraph>
      <Paragraph position="1"> the group opinion and dehvers the result to the next blackboard * Knowledge base The SlmSum knowledge base ts a common knowledge store compnsmg a text representation winch holds all texts in the system and an ontology of the concepts which are needed to deal with them</Paragraph>
    </Section>
  </Section>
  <Section position="5" start_page="91" end_page="91" type="metho">
    <SectionTitle>
3 Computer-oriented discourse representa.
</SectionTitle>
    <Paragraph position="0"> tion Since summarizing is a text and reformation processing task, we have to represent those surface text passages and text meamng umts m the system winch are really worked upon, concentraung on semantic and pragmatic structures The representation must support pragmatic text handhng and deal with bollsUc text structures as well as wtth local rmcrostructures and layout features, because document structure knowledge ts a core item of a professional summanzer's competence * * The practical coding of the vlslble document arcintecture follows SGML conventlons SGML tags like &amp;quot;&lt;hl I&gt; . &lt;hi It:&gt; &amp;quot;Introducuon&amp;quot; &lt;/hl It&gt; &amp;quot;assign a layout feature derived from content structure In the example m table 1, the secUon begmmng Is re&amp;cared by &lt;hi 1&gt; Its Utle is included by the tag pmr &lt;Ill lt&gt; and &lt;/hllt&gt; s The passages that are really read m the simulated working steps are furthermore coded m flrst-order-loglc-hke proposiuonal form (see table 2) Dunng text coding we deliberately chose fitting predicates and standarchzed presentation (e g ordering of arguments, mateinng semanucally nearly eqmvalent words m one concept) Dommn prechcates are dlstlngmshed from predicates that describe an mteracUon between the authors and their readers &lt;hl 1&gt; &lt;hl It&gt; 1 Introduction &lt;/hl It&gt; * &lt;bodyl I&gt; &lt;p&gt; Tins study forms part of the project &amp;quot;Atmogenous and geogenous components m the heavy metal balance of forest trees&amp;quot; The goal of tins project is, on the basis of the distribution within the tree, to trace paths of heavy metal absorpuon and the regularities of their mternal redistribution Furthermore, R ts anned to estimate absorpUon and rechstnbuuon rates In order to obtmn as clear results as possible, the majority of trees analyzed were located m areas with atmogenous or geogenous pollution In conunuatlon of the prewous studies, winch concentrated on trees m contaminated dead ore areas and Black Forest locations with low atmogenous polluUon, the following reports about trees influenced by Ingh atmogenous deposits m the chstnct  While the SGML and the propositional representations are precoded, the discourse level document structures are reconstructed during summarizing The cognitive agents install the respective RST relations Only a few of the most necessary and most simple RST relations have been implemented ELABORATION,</Paragraph>
  </Section>
  <Section position="6" start_page="91" end_page="91" type="metho">
    <SectionTitle>
RESTATEMENT, PURPOSE, CAUSE/
RESULT, EXAMPLE
</SectionTitle>
    <Paragraph position="0"> A small parsimonious ontology has been coded for every document, where the used concepts are organized m a small and very flat hierarchy The ontology is divided into two parts according to Penman (1989) The upper model is domain independent and therefore used for all texts m the system, whereas the lower model is dommn specific, so that one is modelled for each document The agents do some basle lrfferencmg such as comparing text units with knowledge base entries and installing relanons from a fixed set between text umts</Paragraph>
  </Section>
  <Section position="7" start_page="91" end_page="93" type="metho">
    <SectionTitle>
4 Agents
</SectionTitle>
    <Paragraph position="0"> 'The core of the SlmSum simulation are object-onented agents As representatives of the empirically found eogn!tive strategies they manage the reducUon of a large document to &amp;quot; a short summary Agents &amp;ffer m the representations they work upon Some of them are sensitive for SGML tags, others need the propositional presentation to run their methods null In the &amp;msum system, 39 agents are modelled m great detml They are revolved m the central information reduction task of summanzmg, e g the relevance agents Reading and wnting strategies are realized carefully only m so far as they are specific for professional summanzaUon, otherwise they remain black box agents About half of the agents are &amp;quot;real&amp;quot; agents and the rest are &amp;quot;pseudo&amp;quot;agents For instance, the explore agent Is a black box agent of understanding It fakes text comprehension by assigning input passages a precoded propositional representation The reorgamze agent is a black box agent as *well It Is presumed to impose Enghsh grammar and spelling which is not a specific subtask of professional summarizing Therefore the agent functions more or less as a placeholder The agents fall into the following functmnal classes planmng and control, exploration, relevance assessment, target textconstructton, quahty enhancement, formulation, and general knowledge processing In addition, there are rmnor agents such as readers and writers * To make the agents more concrete, we discuss m the following two &amp;quot;real&amp;quot; relevance agents that happen to be good old acqumntances of everybody m automatic summanzmg relevant-texthmt (realizing the indicator phrase method, see, e g, Palce, 1990 and Borko 1968).and relevant-call, which assesses the importance of an entlty by measuring its distance from the theme (pnnciple used m Jacobs &amp; Rau, 1990, .McKeown, 1985, Trabasso &amp; Sperry, 1985) More about agents is found m Endres-Niggemeyer et al (1995) and m Endres-Niggemeyer . (I 997)  Relevance agents work under the control of hold, the responsmble agent for the relevance blackboard, (cf fig 2) Since the skdled red.uct~on of document meaning to the most relevant items ~s central to professional summanzatzon, hold ms m charge of the core of the whole summarizing process * Relevant-texthint The relevant-texthmt agent mmplements the &amp;quot;mdmcator phrase method&amp;quot; known since the early days of automatic abstracting It exploits cue phrases by which authors quahfy their statements, assunung that the quahfica-. tlon apphes to the scope of the indicator phrase By rots mere presence, a (posmve) re&amp;cater phrase, expresses the author's emphasis and suggests the relevance of the statement m its scope In addmon, cue phrases often explain what the author announces, e g a new fin&amp;ng or the content of the conclusmon, and ~ts role m the&amp;quot; document Relevant-texthmt reads the proposltmons on the meaning panel of the document blackboard (see fig 2) To make out relevant propos~tlons, mt uses a private dlcuonary, where the potential mdmcator predmcates (cf.</Paragraph>
    <Paragraph position="1"> table 3) are hsted Since the &amp;cuonary entries are annotated with mterpretaUons, the agent can draw the attention of other agents to these proposmons by passing them parts of its pnvate knowledge . &amp;quot; Relevant-texth~nt recogmzes the mdmcator predmcates by simple pattern matclung as contmmng an. indicator phrase, matching rots dmctaonary entry with a proposmon such as proposmon 5 mn table 2 Consequently, the agent annotates proposmon 4 as describing the project theme and therefore as Important and puts it together wroth others on the relevance blackboard (see fig 2 and table 4) * Relevant-caU Relevant-call recoguizes a text meamng totem as relevant because it hnks it to the document theme (see figure 3) The agent needs the themauc structure and, as a candidate for linkage to the document topmc, a text proposmon The agent checks whether an open RST-type hnk of the document theme is able to attach the candidate If so, the proposmon Is regarded as relevant and added to the document theme Theme-of-document.</Paragraph>
    <Paragraph position="2">  dommn_lnvestlgate (project, X) domam_parUclpate (studyjhls, X) dommn_esumate (project, X, ram) mteracuon_report (author, X) dommn contmue (researchers, Y, X)  extension 1 Figure 3 Relevant-call expands the document theme To find the document theme, relevant-call accesses the theme panel of the document blackboard The agent tries to attach proposmons discovered by other (data-oneuted) agents For instance ~t picks up proposmon 4 recommended by relevant-texthmt because it states what the research ~s about ('This mvest~ganon forms part of a project &amp;quot; - cf table 4) Relevant-call tries all available RST-relataons m order to hnk proposltaon 4 to the document theme (in extenston 1) It ~s easy to see what happens proposmon 4 rephrases the theme, the concepts &amp;quot;pollute&amp;quot;, &amp;quot;heavy_metal&amp;quot;, and &amp;quot;forest_trees&amp;quot; of the theme are repeated The theme and the text proposmon unify, but proposmon 4. bnngs some addmonal mformatton about the (\[geogenous, atmogenous\]) components of contmmnataon This corresponds to an elaborauon of the theme Consequently, the proposRon. Is attached by an ELABORATION hnk Th6 new hypothesis of a topic structure ~s given m figure 3 At that moment, two new proposltaons have been attached tothe theme, so that the theme has three extensions</Paragraph>
  </Section>
class="xml-element"></Paper>