File Information
File: 05-lr/acl_arc_1_sum/cleansed_text/xml_by_section/metho/94/w94-0315_metho.xml
Size: 25,229 bytes
Last Modified: 2025-10-06 14:13:54
<?xml version="1.0" standalone="yes"?> <Paper uid="W94-0315"> <Title>Using a textual representational level component in the context of discourse or dialogue generation</Title> <Section position="4" start_page="128" end_page="134" type="metho"> <SectionTitle> 2 Micro-planning component </SectionTitle> <Paragraph position="0"> In order to solve the above mentioned problems, a component can be added between the message planner and the linguistic realizer (e.g., see \[Levelt 89\] or \[Hovy 92\]). The role of this third component, called the micro-planner in \[Levelt 89\], is to plan the best informational perspective that can be given to the message. The component often uses an intermediate representation level. For example, in the system Epicure \[Dale 88\], a recoverable semantic structure, which specifies the underlying content of the utterance to be generated, and an abstract syntactic structure, which is much closer to the surface syntax of the linguistic form which will be produced, are used.</Paragraph> <Paragraph position="1"> Dale also shows how the two levels of representation play a role in the generation of elisions and one-anaphoric expressions. However, the abstract syntactic level of most generation systems is often hazy and dependent on the syntactic formalism used by the linguistic realization component. null Our intermediate structure, which we call textual 1 structure, is expressed in terms of abstract linguistic resources \[Meteer 92\] 2. These resources are abstractions over concrete resources of language (e.g., syntactic structures, words, and grammatical features); they reflect linguistic constraints while abstracting away from syntactic details. They are useful for generation systems since allowing a generation system to select the concrete resources.., makes available many more degrees of freedom than the language actually permits \[Meteer 92, p. 67\].</Paragraph> <Paragraph position="2"> Abstract linguistic resources become the terms in which the micro-planning component takes its decisions and that way they constrain the system to only do what is expressible in the language since it is not allowed to choose them independently. We made some modifications on two kinds of abstract resources defined by Meteer: we improved her semantic categories by using perspectives from systemic functional linguistics (cf. section 2.1) and we extended the notion of resource trees (cf. section 2.2).</Paragraph> <Paragraph position="3"> l&quot;Textual&quot; is used in the sense of the textual metafunction in the systemic functional theory (see below). For presentations of systemic functional linguistics, see \[Halliday 85\] or \[Matthiessen & Bateman 91\].</Paragraph> <Paragraph position="4"> 2Meteer used the abstract linguistic resources to build a structure which she called text structure. We do not use the same name because Meteer's structure could exist at the same level of rhetorical structure theory (i.e., tile message planning level) whereas our structure is complementary to this level.</Paragraph> <Section position="1" start_page="129" end_page="129" type="sub_section"> <SectionTitle> 7th International Generation Workshop deg Kennebunkport, Maine * June 21-24, 1994 </SectionTitle> <Paragraph position="0"> Moreover, we explicitly defined lezical resources (cf. section 2.3).</Paragraph> </Section> <Section position="2" start_page="129" end_page="131" type="sub_section"> <SectionTitle> 2.1 Semantic hierarchies </SectionTitle> <Paragraph position="0"> Semantic categories (cf. figure 2) reflect pertinent combinations of linguistic features (e.g., syntactic features and lexical items). For example, the five sub-categories of object s are specified from definiteness (i.e., definite or indefinite) and number (i.e., singular or plural). These categories constrain the composition of the utterance by defining what kinds of constituents can be extended and how they can be composed. For example, an adjunct constituent of type manner can be added to an event (e.g., Peter suddenly decided to go to the beach) whereas an adjunct of type property cannot (e.g., *Peter importantly decided to go to the beach4); a process-event can be composed with a durative-adverbial (e.g., Karen watched TV for an hour) whereas a transition-event may be composed of a frame-adverbial (e.g., Karen solved the problem in one hour).</Paragraph> <Paragraph position="1"> On the one hand, we think that according to systemic linguistic theory s, the hierarchy of semantic categories mikes ideational constraints and textual constraints; both are semantic, but semantic of rather different kinds. Thus, while it is correct that the categories event, manner, and property are semantic (ideational) categories, taking an event perspective on something is a textual decision, e.g., &quot;The ship sank.&quot; is an ideational event, and it is textually presented within an EVENT-PERSPECTIVE.</Paragraph> <Paragraph position="2"> &quot;The sinking of the ship&quot; is still an ideational event, but it is now presented within an OBJECT-PERSPECTIVE. Even clearer are cases such as definite and indefinite noun phrases. Thus the differences between &quot;a man&quot; and &quot;the man&quot; can be characterized as follows. They are both ideational of category PERSON. The noun phrase '% man&quot; presents a discourse entity of that ideational semantic category which is textually UNIDENTIFIABLE. This is a category drawn from textual semantics. Textual semantic classifications of discourse entities change as a text develops. The noun phrase &quot;the man&quot; is classified textually as IDENTIFI-ABLE. The former is realized grammatically (largely on sin this theory, language is organized functionally according to three metafunctions, concerned with the representation of experience (ideational), symbolic interaction between interlocutors (interpersonal), and presentation of information as text in context (textual).</Paragraph> <Paragraph position="3"> the basis of the textual semantic classification) as an indefinite noun phrase, the latter as a definite noun phrase. In this way, ideational semantics and textual semantics are complementary to each other and both help constrain grammatical realization.</Paragraph> <Paragraph position="4"> On the other hand, we think that the textual part of Meteer's hierarchy is too abstract to treat phenomena that need syntactic information such as anaphora planning and selection of grammatical constructions (see also the other examples of linguistic phenomena given in section 1); in particular, syntactic information such as gender or verb voice is missing.</Paragraph> <Paragraph position="5"> In order to solve these problems, we propose to re-place the hierarchy with two constituent hierarchies: an ideational semantic hierarchy and a textual semantic hierarchy. The former is based on the Upper Model defined in \[Bateman et al. 90\]. The latter was developed by ourselves; we call it the hierarchy of textual semantic categories. These two hierarchies are not incompatible since \[a\] language.., embodies a multiplicity of constituent hierarchies, coexisting in different parts of the system. They are not unrelated: on the contrary, lhey are all different facets of the same phenomenon. \[...\] At the same time, each of these hierarchies is independent of the others, and lhere are infinite possibilities of matching them up in a meaningful way \[Halliday 85, p.</Paragraph> <Paragraph position="6"> 18\].</Paragraph> <Paragraph position="7"> The Upper Model provides a domain-independent, general and reusable conceptual organization which is used to classify the world knowledge when linguistic processing is to be performed. This general semantic taxonomy can also be seen as an inheritance hierarchy that organizes concepts according to how they can be expressed (cf. figure 3). The Upper Model also'expresses (ideational) semantic restrictions on arguments required by entities it involves. For example, the realization-event 6 requires two participants: an agent that must be an animate and a patient that must be a product. The Upper Model also provides the type of qualities which may be ascribed to particular concepts. For example, the Upper Model expresses that the destination relation relates a place object to an event.</Paragraph> <Paragraph position="8"> The hierarchy of textual semantic categories is a property inheritance network that represents textual knowledge of the target language. Figure 4 shows a part of our hierarchy; the whole hierarchy contains about eighty categories. The definition of the textual semantic categories is based on an analysis of French \[Bescherelle 90\] and of English \[nalliday 85, Shinghal 92\].</Paragraph> </Section> <Section position="3" start_page="131" end_page="131" type="sub_section"> <SectionTitle> 7th International Generation Workshop * Kennebunkport, Maine * June 21-24, 1994 </SectionTitle> <Paragraph position="0"> We will explain the meaning of each textual category by describing the way in which they are realized from a grammatical viewpoint. Complex-clauses 7 are realized as sentences. The sub-hierarchy of clause represents different linguistic forms of clause; in particular, the timeanchored-event is realized as a finite clause, the activity as a non finite clause and the object-perspective as clause nominalization. The clause-nucleus sub-hierarchy classifies different kinds of verbs (e.g., transitive and intransitive verbs) and verb voice. The other-argument is realized as verb argument that has no specific grammatical function, whereas the three other clause-nucleus-arguments are realized as constituents whose grammatical function is subject, direct object or indirect object. For example, the verb &quot;to get to&quot; requires two arguments: one subject-argument (realized as the subject) and one other-argument which represents the destination. The textualobject is realized as a noun phrase, and, in particular, the unidentifiable-object as an indefinite noun phrase and the identifiable-object as a definite one. The hierarchy of textual-object-nucleus corresponds to nouns; it represents various linguistic forms of nouns with respect to their gender and number. The modifiers are realized as adjectives, adverbs, prepositional phrases, and subordinate clauses.</Paragraph> </Section> <Section position="4" start_page="131" end_page="131" type="sub_section"> <SectionTitle> 2.2 Resource trees </SectionTitle> <Paragraph position="0"> Resource trees are primitive trees from which textual structures are built. In particular, they indicate the way in which a constituent (i.e., a textual structure node) can be expanded, that is, what are its sub--constituents, or exgended, that is, what kind of additional sub-constituents can be added. They are defined from elementary trees.</Paragraph> <Paragraph position="1"> Meteer distinguished two basic forms of elementary trees: the kernel tree and the composite tree. These two trees represent Halliday's distinction between the multivariate and the univariate structure, respectively \[Halliday 85\].</Paragraph> <Paragraph position="2"> The kernel tree (i.e., the HEAD-ARGUMENT structure) is an atomic structure, that means, it cannot be modified independently from the whole structure. Essentially, it represents a predicate/argument structure. The composite tree is an incremental structure, that is, new child nodes can be added at any time of the expansion of the mother node. Meteer distinguished the COMPOSITE-MATRIX-ADJUNCT structure, which corresponds to the hypotactic relation (i.e., the relation between dependent elements and their dominating node), and the COMPOSITE-COORDINATE structure, which represents the paratactic relation (i.e., the relation between elements of equal status). Meteer defined resource trees as follows: This definition expresses that the complex-event resource tree is a COMPOSITE tree whose MATRIX has to be of type state or event and ADJUNCTs of type raarmer or location. It also specifies a constructor function which builds the tree.</Paragraph> <Paragraph position="3"> We think that the language for defining resource trees is not powerful enough. For instance, it must be explicitly specified that a composite tree only has one matrix node or that an activ+-~y does not have an explicit subject. In order to solve this problem, a more powerful language had to be defined. In particular, the restriction on daughter nodes is more precise. In addition, we introduce the notions of relevance and constraints. Restrictions are expressed by using a specific language which expresses quantities (e.g., &quot;one&quot; or &quot;several&quot;) and description of the daughter constituents. In particular, each daughter constituent is described by a structure, which we call abstract constituent, which specifies the ideational and textual semantic categories of the constituent, its structural relationship with the mother node, and its relevance. The relevance is the link between textual structure nodes and parts of the message; it indicates the part of the message verbalized by the mother node and its child nodes. Constraints are conditions which must be satisfied by the textual structure. For example, constraints can express that the subject-argument of an infinitive-activity will not explicitly appear in the final text. Figure 5 shows the definition of the &quot;infinitive-activity&quot; resource tree s, which is a composite tree whose mother node has the textual category infinitive-activity. The tree has one and only one matrix node of the ideational category process and of the textual category clause-nucleus, and additional adjunct nodes of the ideational category circumstance and of the textual category clause-modifier. It specifies a constraint which states that the subject-argument must be implicit. It also indicates that the mother node and its daughter nodes correspond to the same part of the message, that is, they have the same relevance (cf.</Paragraph> <Paragraph position="4"> the variable X in figure 5) 9.</Paragraph> </Section> <Section position="5" start_page="131" end_page="132" type="sub_section"> <SectionTitle> 2.3 Lexical resources </SectionTitle> <Paragraph position="0"> Lexical items, i.e., words, are lexical resources of the language. We represent their meaning by associating each of them with entities of the Upper Model and of the tex8Meteer's resource trees axe expressed in terms of semantic categories whereas our resource trees axe defined in terms of textual categories.</Paragraph> <Paragraph position="1"> 9 The variable of the field &quot;relevance&quot; is unified with the relevance of a textual structure's node during the expansion of the node with the resource tree.</Paragraph> <Paragraph position="2"> tual semantic hierarchy. For example, the lexical resource which corresponds to the direct transitive verb &quot;to write&quot; is defined as follows:</Paragraph> <Paragraph position="4"> A lexical resource can be linked to more than one textual category; for example, the following resources</Paragraph> <Paragraph position="6"> correspond to the nouns &quot;article&quot; and &quot;articles&quot; and to the adjective &quot;quick&quot; and the adverb &quot;quickly&quot;, respectively. A lexical resource can also be associated with several ideational categories; for example, the following</Paragraph> <Paragraph position="8"> can be used to verbalize a rhetorical relation &quot;purpose&quot; or &quot;enablement&quot;.</Paragraph> </Section> <Section position="6" start_page="132" end_page="134" type="sub_section"> <SectionTitle> 2.4 Textual structure </SectionTitle> <Paragraph position="0"> The textual structure is a tree-like structure built from resource trees. It represents the constituency of the utterance (each node is a complete constituent of the utterance), the structural relations among constituents (head-argument, composite-matrix-adjunct, compositecoordinate), the ideational and textual categories, the relevance and the restriction of each node, and the lexemes.</Paragraph> <Paragraph position="1"> In addition, a marker which describes the lexical cohesion of the text (see \[Halliday & Hasan 76\]) is associated to each lexeme. The textual structure provides linguistic information which is required for treating specific linguistic phenomena such as anaphora planning (see section 1).</Paragraph> <Paragraph position="2"> For example, in figure 6 (only the main information contained in each node is depicted), the textual structure indicates that the domain instance &quot;Stock&quot; corresponds to a male and will be verbalized as a singular masculine proper name (named-object + singular-masculine-name).</Paragraph> <Paragraph position="3"> Several textual structures can be built to verbalize one message. The structure which is selected is the one which has the higher cohesive score. This score is obtained by summing the scores of lexical cohesion of all lexical re- null context of dialogue A plan of dialogue acts is verbalized by transforming it into two successive representational levels: surface linguistic acts and textual representations. These two main transformations are achieved with respect to the current dialogue context. We represent the context through four histories. The history of dialogue acts stores, in particular, the requests of the interlocutor that have not been</Paragraph> </Section> <Section position="7" start_page="134" end_page="134" type="sub_section"> <SectionTitle> 7th International Generation Workshop * Kennebunkport, Maine deg June 21-24, 1994 </SectionTitle> <Paragraph position="0"> answered yet. Each item of this list provides information on the dialogue act itself, the way in which conceptual entities that appear in its propositional content were realized with respect to the lexical semantics and the list of linguistic resources used. This information constitutes a local context during the verbalization of the answer.</Paragraph> <Paragraph position="1"> The history of conceptual entities contains information on the conceptual entities (e.g., &quot;arrival-airport&quot;) already verbalized in the dialogue and the way in which they were realized with respect to lexical semantics. This history is used for the verbalization of all surface linguistic acts, except for &quot;answers&quot; (the semantic realization of conceptual entities that appear in an &quot;answer&quot; is constrained by the local context retrieved from the history of dialogue acts). The history of objects involves information on the instances of conceptual entities (e.g., &quot;JFK&quot;) mentioned in the dialogue. It is used to retrieve objects that are required from the lexical semantics point of view during the surface linguistic act planning (cf. the ideational semantic restrictions on arguments of entities of the Upper Model). It also provides information that is useful during anaphora planning, such as the information status (given or new) of objects. Finally, the history of linguistic-resources indicates which abstract linguistic resources have been used in the dialogue. It is divided into two sets: an active set and an open set. The active set contains the linguistic resources used in the last sentence of the dialogue, and the open set the resources used in the preceding sentences. In particular, this history is used to evaluate the cohesion of textual structures.</Paragraph> <Paragraph position="2"> We will take an example to show how the generation system deals with these histories in order to verbalize a plan of dialogue acts. We will assume that the interlocutor (i.e., the user) uttered the following question during the dialogue: The 10:00 flight goes to which airport? The dialogue act planner builds a dialogue act of informing that the airport is &quot;JFK &quot;1deg (this act is shown in figure 1). Since there is a request in the history of dialogue acts that corresponds to the dialogue act, the surface linguistic act planner decides to build an &quot;answer&quot; surface linguistic act. Then, it uses the local context provided by this history in order to transform the semantico-conceptual content into a semantic content. In particular, it builds a propositional content expressing a destinationevent whose theme is &quot;the 10:00 flight&quot; and whose destination is &quot;JFK&quot; ~cf. content 1 in figure 1). Finally, the constructed linguistic act is passed to the micro-planning component. This component builds at least two textual structures representing the cases where the des~inationeven~ is verbalized by &quot;to go&quot; or &quot;to arrive&quot;. Since &quot;to go&quot; appears in the local context, the selected textual structure is the one corresponding to the following utterance: The 10:00 .flight goes to JFK.</Paragraph> <Paragraph position="3"> According to the content of the history of objects, the micro-planner can introduce anaphoric expressions. In that case, the built textual structure corresponds to: It goes to JFK.</Paragraph> <Paragraph position="4"> Finally, using the history of linguistic resources, the planner can produce ellipsis, in which case, the textual structure built corresponds to: To JFK.</Paragraph> </Section> </Section> <Section position="5" start_page="134" end_page="135" type="metho"> <SectionTitle> 4 Concluding remarks </SectionTitle> <Paragraph position="0"> In this paper we described a component of a natural language generation, called the micro-planner, which is situated between the message planner and the linguistic realization component. Its role is to bridge the gap between the two traditional components in order to avoid some of the problems which arise in two-component generation systems. In order to achieve this, the micro-planner transforms the message built by the message planner into an intermediate structure, which we call textual structure. This structure represents an abstract grammatical organization of the final text. In particular, it provides semantic and syntactic information, which is useful to handle linguistic phenomena such as anaphora planning.</Paragraph> <Paragraph position="1"> This new representational level is based on the notion of abstract linguistic resources \[Meteer 92\]. We made two main modifications. Firstly, Meteer's semantic category hierarchy was substituted by two distinct semantic hierarchies: an ideational semantic hierarchy (i.e., the Upper Model \[Bateman et al. 90\]) and a textual semantic hierarchy. Secondly, we increased the expressibility of the resource tree definition language. Moreover, we defined a simple mechanism to represent lexical resources of the language.</Paragraph> <Paragraph position="2"> In a three-component generation system, the micro-planner makes the linguistic decisions (e.g., the selection of syntactic structures and the choice of lexical items) that are usually taken by the linguistic realizer in a two-component generation system. The advantages are, firstly, taking a decision is simpler since the micro-planner de~ls with abstract linguistic knowledge, secondly, the micro-planner is closer to the message planner than the linguistic realizer, and, thirdly~ the linguistic realization component is simplified since it no longer has to take linguistic decisions. Such a component appears in the generation system of \[Danlos 87\]: the strategic component takes conceptual and linguistic decisions, and the tactical component only applies grammar rules. In fact, the</Paragraph> </Section> <Section position="6" start_page="135" end_page="135" type="metho"> <SectionTitle> 7th International Generation Workshop * Kennebunkport, Maine deg June 21-24, 1994 </SectionTitle> <Paragraph position="0"> linguistic realization component only deals with syntactic details such as agreement and complete surface order and morphology.</Paragraph> <Paragraph position="1"> Our micro-planning component is currently being integrated in a text generation system developed at IRST (see \[Maier & Not 93, Panaget 94\]). A previous version (\[Panaget 92\]) was implemented in order to be integrated in a human-computer oral dialogue systefn developed at CNET \[Sadek et al. 94\]. For the former case, the linguistic realization component \[Pianesi 91\] is based on the government and binding theory and a head-driven bottom-up generation algorithm. The component consists of an integrated system for editing, modifying and maintaining unification-based grammars. For the latter case, the linguistic realizer is based on Tree Adjoining Grammar \[Joshi 87\]. In particular, the textual structure is transformed into a derivation tree (i.e., a tree which specifies a derived tree through a bottom-up traversal), which is then 'executed' by the linguistic component in order to build a derived tree (i.e., a syntactic tree).</Paragraph> </Section> class="xml-element"></Paper>