File Information
File: 05-lr/acl_arc_1_sum/cleansed_text/xml_by_section/metho/99/w99-0108_metho.xml
Size: 25,028 bytes
Last Modified: 2025-10-06 14:15:25
<?xml version="1.0" standalone="yes"?> <Paper uid="W99-0108"> <Title>Q @ @ @ O O O @ @ O @ O O @ @ 0 @ @ O @ O @ O O O @ @ @ O @ O O @ @ O O @ @ O @ O O O O Generating Anaphoric Expressions: Pronoun or Definite Description?</Title> <Section position="5" start_page="64" end_page="64" type="metho"> <SectionTitle> 3 Long- and Short-Distance Anaphor/c </SectionTitle> <Paragraph position="0"/> <Section position="1" start_page="64" end_page="64" type="sub_section"> <SectionTitle> Expressions </SectionTitle> <Paragraph position="0"> In studying naturally oceu~ng texts in order to identify patterns of pronoun usage, we found consistent patterns over both long distances (i.e., where the last reference to the entity being referred to was more than two sentences back in the discourse) and over short distances (i.e., where the last reference to the entity being referred to was in the current sentence).</Paragraph> <Paragraph position="1"> In long-distance situations we have found that a definite description is almost always used. In short-distance situations a pronoun is almost always used (except in situations where this pronoun is ambiguous). Thus, the sentence seems to be a very important construct to consider in choosing anaphoric expressions.</Paragraph> <Paragraph position="2"> We now turn to consider the factors that might be affecting the expression choice in situations other than these. We hypothesize that discourse structure must be considered for such cases.</Paragraph> </Section> </Section> <Section position="6" start_page="64" end_page="66" type="metho"> <SectionTitle> 4 &quot;lime-Threaded Discourse Structure </SectionTitle> <Paragraph position="0"> In this section, we describe our approach to using dis.</Paragraph> <Paragraph position="1"> course structure for choosing the fight referring expression. Since we are working with stories from hewspapers we were not able to identify the kind of discourse structure as assumed by Orosz & $idner (1986), whose dialogues are more task-oriented and have clear intentional goals. Instead, the texts have a structure consisting of multiple story lines which we call threads (cf. also Ro~ et al. (|995)). A thread describes a particular part of the story. It can be interrupted by other threads and continued later. Thus the tiueaded structure is more complicated than the hierarchical (~ee-like) structure posited in Cn~sz & Sidner (1986) and Mann & Thompson (1988).</Paragraph> <Paragraph position="2"> Therefore, we do not think that the texts we looked at can be analyzed using a stack (for example).</Paragraph> <Paragraph position="3"> We needed to find a structuring device that was part of the input to a sentence generation system and that was recognizable on the surface (so that we could evaluate our algorithm on naturally occurring text). After invest/ null gating some work on narrative structure (Genette, 1980; Prince. 1982; Vogt, 1990), &quot;we determined that changes in the deictic center of the story (Nakhimoysky, 1988: Wiebe, 1994) not only must be part of the input to a sentence generator (e.g., for appropriate tense generation), but were also both well marked in the text and seemed to have an influence on anaphoric expression choice.</Paragraph> <Paragraph position="4"> A shift in the deictic center can be signaled by a shift of topic, a shift of time scale, a shift in spatial scale, or a shift in perspective (Nakhimovsky, 1988; Wiebe, 1994). Since a shift in time scale is often indicated by linguistic means and the time being referred to must be part of the input to a sentence generator, we concentrated on this point. We also acknowledge that the changes in time seem quite important in the news stories that we analyzed. Other genres of text might depend on other kinds of structuring devices.</Paragraph> <Paragraph position="5"> Changes in time scale or time, as we redefined the category, may require world knowledge reasoning to recoguize but are often indicated by either cue words and phrases (e.g., &quot;n/he years ago &quot;, &quot; a year&quot;, &quot;for months'; &quot;several months ago&quot;), a change in .grammatical time of the verb (e.g., past tense versus present tense), or changes in aspect (e.g., atomic versus extended events versus states as defined by Moens & Steedman (1988)).</Paragraph> <Paragraph position="6"> In considering how time change might affect anaphoric expression choice, we consider the choice for the first mention of a discourse entity in a sentence where that entity has recently been referred to in the discourse.</Paragraph> <Paragraph position="7"> Our hypothesis is that: Changes in time reliably signal changes of the thread in newspaper articles; definite descriptions should appear when the current reference to a discourse entity is in a different thread from the last reference to that entity and pronouns should occur when the previous mention is in the same thread 3.</Paragraph> <Paragraph position="8"> In order to evaluate this hypothesis, we mapped out the time being referenced in our texts on a clause-byclause basis. For each clause in the texts we indicated the time which was referred to. We distinguished between events that occurred at a single instance in time (atomi c events) and events or states that occurred over a span of time (repeated atomic events, extended events, and states). For atomic events we allowed for both a specific time at which it occurred and for a non-specific time that indicated the range of uncertainty. We allowed time spans to have both specific end points and unspecified end points as well.</Paragraph> <Paragraph position="9"> An example from our corpus with its associated temporal structure may illustrate these labels, the complexity of the texts under consideration, and how we propose pronoun generation is affected.</Paragraph> <Paragraph position="10"> 3Note that tbe discourse thread maychange between the two references in question. This would be signaled by a change in time in e_laeses between the two references in question. We are interested in whethor the two references are in the same (use a pronoun) or different (use a definite description) thread.</Paragraph> <Paragraph position="11"> Example 2: (47a) Questioned about the criminal activities of the football club, (47b) Mrs. Mandela maintained (47c) that she had never had any control over them.</Paragraph> <Paragraph position="12"> (48a) This despite testimony from a half dozen former members (48b) that they even had to get permission to go in and out of her yard.</Paragraph> <Paragraph position="13"> (49a) Mrs. Mandela also said (49b) she had disbanded the club (49c) after her husband asked her to, despite evidence to the contrary.</Paragraph> <Paragraph position="14"> (50) Mrs. Mandela faced questions from more than 10 lawyers representing various victims and the panel of commissioners and their investigators.</Paragraph> <Paragraph position="15"> ............. ....... deg.-.degoo o.o..&quot; ......... I .... i ....</Paragraph> <Paragraph position="16"> . .............. .:. ......... , .............. ~.....</Paragraph> <Paragraph position="17"> **.deg., ..... ..oedegodeg,.odegdeg.~..degdeg ...... ..o.:..deg oo, a' D' el a' b a' b' el I 47 48 49 50 Notice that sentence 47 consists of three clauses. The first two (47a and 471)) describe atomic events that are taking place at the &quot;now&quot; time of the story (during the proceedings against Mrs. Mandela). The third clause (47c) refers to an indefinite span of time in the past (during which Mrs. Mandela's football club existed). Note, the use of the past perfect in (47c) indicating the change in time and setting.</Paragraph> <Paragraph position="18"> * As Figure I illustrates, there is a name (N) reference to Mrs. Mandela in (47b), and a pronominal 0P) reference -to her in (47c). This pronoun is used even though there is a change in time between (47b) and (47c) and is explained because this is a sho~-distance reference since it is the second reference to Mrs. Mandela within the same sentence (a condition that overrules the time change hypothesis). null (48a) represents a change hack to the time of the proceedings (note the discourse-deictic reference (Wehber, 1991) &quot;This&quot;). (48b) again points to the time in the past, though this is not explicitly marked linguistically as it was in (47c). Here world knowledge must be used to understand the time referenced in this clause. Note, however, that the time would have to be part of the input to a generator, and thus our rules are completely well-defined from the generation perspectiv e .</Paragraph> <Paragraph position="19"> The use of a pronoun to refer to Mrs. Mandela in (48b) is warranted by our hypothesis, because the previous mention of Mrs. Mandela in (47c) references the same time as is referenced in (48b). &quot; Because there is a time change between (48b) and (49a), our hypothesis explains the appearance of the proper name in (49a) even though it occurs just after a pronoun (in (48b)) co-specifying the same character? The pronouns in the remainder of(49) are explained because they are subsequent references within the same sentence (despite the fact that they refer to an unspecified time in the past which is different from the time referenced in (49a)). Finally, theuse of a name in (50) is again indicated by the change in time between (49c) and (50).</Paragraph> </Section> <Section position="7" start_page="66" end_page="67" type="metho"> <SectionTitle> 5 Ambiguities </SectionTitle> <Paragraph position="0"> Of course, the choice of referring expression is not only guided by discourse structure, there is also an influence due to ambiguities. Dale (1992) generated referring expressions so that their referents could be distinguished from the other discourse entities mentioned in the context. This strategy can be interpreted as: Generate a pronoun whenever it is not ambiguous. However, how one should define context is not quite clear. For this definition we choose a span of text considered important in our previous work on anaphora resolution (Slrube, 1998), and define a referring expression as ambiguous if there is a competing antecedent (i.e. another discourse entity matching in number and gender) mentioned in the previous sentence or to the leR of the referring expression in the current sentence. Of the 437 referring expressions in the texts we analyzed, 104 were considered ambiguous by this definition. Of these only 51 were realized as a definite description. Thus a rule which specifies use of a definite description if a pronoun would be ambiguous according to this definition appears to be too strict. Therefore we need to consider ambiguous cases in more detail.</Paragraph> <Paragraph position="1"> sNote that the use of this definite description cannot be explained by a topic shift since there is no topic shift in between the previous text and (49). At least two discourse entities (&quot;Mrs. Mandela&quot; and the &quot;football club&quot;) are constant, only Mrs. Mandela's husband does not occur in the immediately preceding sentences.</Paragraph> <Paragraph position="2"> Consider the following excerpt taken from later in the same story that contains Example i. For convenience.</Paragraph> <Paragraph position="3"> references to the main character appear in bold, while those to a competing antecedent appear in italics.</Paragraph> <Section position="1" start_page="66" end_page="67" type="sub_section"> <SectionTitle> Example 3: </SectionTitle> <Paragraph position="0"> Mr. Curtis might have lived out his live in obscurity if it were not for a New Haven television reporter. Jim Hoffer, of station WTNH.</Paragraph> <Paragraph position="1"> who got an anonymous tip last summer that Mr. Curtis had been attending college. Using a hidden camera, a crew taped Mr. Curlis going to his classes at Southern Connecticut State University, and a producer recorded a conversation with him in a cafeteria. In a clear, steady voice, Mr. Curtis can be heard to say that he...</Paragraph> <Paragraph position="2"> Concentrating on the references to Mr. Curtis, we question when a pronoun can be used, and when the name should be used. To handle cases where competing antecedents occurred, we turned to pronoun resolution algorithms. Our intuition was that a pronoun could be generated to refer to a particular discourse entity if * a pronoun resolution algnr/thin woul d choose that entity as a referent for the pronoun. To our knowledge, there are only two focus-based pronoun resolution algorithms that are specified in enough detail to work on unrestricted naturally occurring text: Brennan et al. (1987) using the definition of utterance according to Kameyama (1998), and Struhe (1998). Strube evaluated the effectiveness of these two algorithms on the task of pronoun resolution in some naturally occurring texts. Because Strube's algorithm showed significantly better results, we have turned to it for guidance in pronoun generation.</Paragraph> <Paragraph position="3"> The idea is that if we want to refer to a discourse entity, E, but there is a competing antecedent, C, we look to Strube'S algorithm in the following way. If Strube's algorithm would resolve a pronoun to be K we use a pronoun. If, instead, Strube's algorithm would prefer C as the referent of the pronoun, we would use a definite description to refer to ~-.</Paragraph> <Paragraph position="4"> We evaluated this idea along with several other alternatives (e.g., using discourse structure alone, using longand short-distance rules ignoring the ambiguity, distingnishing between first and subsequent reference within a sentence, using a definite description whenever there was ambiguity) on the ambiguous examples in our corpus. Our analysis showed that the use of Strube's algnrithm showed improvement, but it seemed to be too liberal with suggesting pronouns when the competing antecedent was in the previous sentence. Assuming Strube's algorithm reflects human processing of referring expressions (something NOT claimed), what this means is that in our texts the writer chose to generate a definite description even though a pronoun would have been informafionally adequate to resolve the referent correctly. This occurred most frequently over sentence boundaries.</Paragraph> <Paragraph position="5"> When a competing antecedcnt is within a sentence, however, Strube's algorithm appears to be quite effective. * Thus the rule we .settled on acknowledged the importance of sentence boundaries and is shown in Figure 2. I. if this is the first occurrence of X in the current sentence and (a) /./'there is a competing antecedent in the previous sentence, use a definite description; (b) /f there is a competing antecedent in the same sentence (i.e.. to the left) and i. if Struhe's algorithm would resolve a pronoun in this position to be X. use a pronoun; ii. e/se use a definite description; 2. /./'this is a subsequent occurrence of X in the current sentence and (a) U&quot; there is an intervening competing antecedent, use a definite description; Co) ~ there is no intervening competing ant_ _,y~,,tttttJent, use a pronoun.</Paragraph> </Section> <Section position="2" start_page="67" end_page="67" type="sub_section"> <SectionTitle> Exist 6 Anaphoric Referring Expression Generation Algorithm </SectionTitle> <Paragraph position="0"> In the previous sections we have argued that in some instances of anaphoric expression choice, the threaded discourse structure must be taken into account. For the particular texts that we analyzed, we argue that threads defined in terms of the time referenced in a clause are appropriate for use. Other kinds of discourses will also exhibit a threaded structure, but the threads themselves might be defined by different means. In addition to discourse structure, the pote, ndal for ambiguity must be considered. Additionally, cases of long- and short-distance anaphora should be handled indepandantlyof the threaded discourse slructure considerations.</Paragraph> <Paragraph position="1"> . Based on these findings, we propose the algorithm for realizing anaphoric expressions shown in Figure 3. Note in this algoriflun we refer to the notion of a discourse thread which might be defined differently for different kinds of texts.</Paragraph> </Section> </Section> <Section position="8" start_page="67" end_page="67" type="metho"> <SectionTitle> 7 Empirical Data </SectionTitle> <Paragraph position="0"> We applied the algorithm described in the previous section to three texts from the New York Tunes. Articles.</Paragraph> <Paragraph position="1"> ranged from a frontpage article *to local news. We applied the algorithm to all references to persons in these texts. The algorithm was correct in 370 cases (84.7%), and wrong in 67 cases (15.3%). In Figure 4 we show the distribution over the rules specified in the algorithm.</Paragraph> <Paragraph position="2"> In order to interpret the results of the algorithm, we must have some comparison. We use a simple scheme which does not consider threaded discourse structure and which bandies ambiguous cases very conservatively.</Paragraph> <Paragraph position="3"> I. If this is a long distance anaphoHc reference (i.e., if the previous reference to X was more than two sentences prior) use a definite description;</Paragraph> </Section> <Section position="9" start_page="67" end_page="67" type="metho"> <SectionTitle> 2. else </SectionTitle> <Paragraph position="0"> if this is an unambiguous reference (i.e., there is rio competing ang~_~_~t) and this is an lntrasentential anaphor (i.e., this is not the first mention of X in the current sentence) use a pronoun;</Paragraph> </Section> <Section position="10" start_page="67" end_page="67" type="metho"> <SectionTitle> 3. else </SectionTitle> <Paragraph position="0"> if this is a thread change (i.e., the previous reference to X occurred in a thread different from the one in which the current reference occurs) use a definite description;</Paragraph> </Section> <Section position="11" start_page="67" end_page="67" type="metho"> <SectionTitle> 4. else </SectionTitle> <Paragraph position="0"> /f there is a competing antecedent (i.e., another object in the previous or current sentence that matches the type and number of X) use the rule found in Figure 2; 5.. else for the remaining cases (i.e., unambiguous cases when time remains the same) use a pronoun. ring Expression for Discourse Entity X This scheme is shown in Figure 5 for comparison purposes. null The results of applying these rules give 343 correct cases (78.5%) and 94 incorrect ones (21.5%). Hence our algorithm reduces the error rate by 28.9%.</Paragraph> </Section> <Section position="12" start_page="67" end_page="68" type="metho"> <SectionTitle> 8 Related Research </SectionTitle> <Paragraph position="0"> A significant amount of work in linguistics has investigated the use of different kinds of anaphoric referring expressions in discourse and their relationship to ease of comprehension. See Arnold (1998) fora discussion of several of the factors involved in refem'ng expression * choice. In many cases the various factors seem to af- null I. /f this is a long distance anaphoric ~ refexence (i.e., if the previous reference to this item is greater than two sentences prior) use a definite description;</Paragraph> </Section> <Section position="13" start_page="68" end_page="68" type="metho"> <SectionTitle> 2. else </SectionTitle> <Paragraph position="0"> /./'there is a competing antecedent (i.e., another object in the previous or current sentences that matches the pronoun which would be used to refer to this entity) use a definite description: 3. e/se the anaphoric expression would be unambiguous so use a pronoun.</Paragraph> <Paragraph position="1"> fect the accessibility of a referent (where accessibility is intended in a broad sense to cover both &quot;topic accessibility&quot; (Givon, 1983) and accessibility due to factors such as recency of mention). Basically, the more accessible a referent the more underspecified a refen'ing expression should be. Accessibility explains the apparent &quot;name-name penalty&quot; as examined in Gordon & Henchick (1998), for example.</Paragraph> <Paragraph position="2"> Our work argues that factors beyond accessibility must be considered in anaphoric expression choice. It is consistent with work such as Vonk et ai. (1992) whose experiments indicate that a referring expression &quot;... that is more specific than is necessary for the recovery of the intended referent ... marks the beginning of a new theme concerning the same discourse referenC'(Vonk et al..</Paragraph> <Paragraph position="3"> 1992, page 304). They argue that such overspecified expressions are serving a discourse function of indicating boundaries. This work does not define what a discourse segment boundary actually is. On the other hand, using the definition of time change as a boundmy condition, our work is consistent with their hypothesis. Interestingly, Vonk et al. (1992) found that in discourses where a theme change was wen marked by other means (e.g., by a preposed adverbial phrase or a subordinate clause indicating time or place) that pronouns were much more common even though a new theme was begun. Presumably such phrases mark the theme change well, and thus it is not necessary to also mark the change via an overspecified description.</Paragraph> <Paragraph position="4"> Approaches which define discourse segments on the basis of reference resolution (Sidner, 1979; Suri & Mc-Coy, 1994; Strobe & Hahn, 1997) are not useful for our purposes because they require referring expressions for recognizing segment boundaries. In contrast to these approaches, we define segment boundaries independently from reference resolution so that in this respect our work is in line with Grnsz & Sidner's (1986) definitions.</Paragraph> </Section> <Section position="14" start_page="68" end_page="68" type="metho"> <SectionTitle> 9 Future Directions </SectionTitle> <Paragraph position="0"> In analyzing our data, there are several places for further consideration. One problem is that our rule which indicates a definite description should be used in a time change overgenerates definite descriptions. Following Vonk et al. (1992) we plan to investigate whether definite descriptions might best be viewed as boundary markers and whether other markers of discourse boundaries (e.g., preposed adverbial phrases) are found in places where our algorithm suggests a definite description because of a time change but a pronoun appear~ in the text.</Paragraph> <Paragraph position="1"> In addition to evaluating more texts under this scheme, there are several places where we will attempt to tighten our methodology. One of these is in the time analysis.</Paragraph> <Paragraph position="2"> Our current analysis distinguishes between four types of time and is driven by both semantic cues in the text (e.g., adverbial time phrases) and changes in tense. Nakhimovsky (1988) also uses changes in &quot;`time scale&quot; as a marker for changes in time. We plan to investigate this to see whether it explains more of the examples. Nakhimovsky also describes several other markers for a setting change, and these will also be investigated to see if they are indicative of definite description use.</Paragraph> <Paragraph position="3"> Another line of future research involves further investigation of the ambiguous cases. Our current rule was developed by evaluating several different possibilities (e.g., using time change rules, different pronoun resolution algorithms) and selecting a rule that explains most of the cases. Still, the number of ambiguous cases is fairly small and analyzing more texts and concentrating on cases where the current rule makes an incorrect prediction may lead us to a more robust rule.</Paragraph> <Paragraph position="4"> Finally, further thought must be put into evaluating the algorithm. In particular, our current evaluation methodology presupposes that the human writer has chosen the best anaphoric expression. It may be interesting to see if there are differences between reading time, eye movements, or comprehension of stories with the human-generated expressions and those using our algorithm.</Paragraph> <Paragraph position="5"> Thought must be put into setting up such experiments and into interpreting results. For example, a faster reading time may not indicate a better choice of referring expression (since the writer may have been interested in an effect other than ease of comprehension). On the other hand, such experiments have the potential for producing a more adequate evaluation of the methodology than does mimicking the human-produced text and should be looked into further.</Paragraph> </Section> class="xml-element"></Paper>