Discourse Structure for Context Question Answering 
Joyce Y. Chai Rong Jin 
Department of Computer Science and Engineering 
Michigan State University  
East Lansing, MI 48864 
jchai@cse.msu.edu, rongjin@cse.msu.edu 
 
 
 
Abstract 
In a real-world setting, questions are not 
asked in isolation, but rather in a cohesive 
manner that involves a sequence of related 
questions to meet user’s information needs. 
The capability to interpret and answer 
questions based on context is important. In 
this paper, we discuss the role of discourse 
modeling in context question answering.  In 
particular, we motivate a semantic-rich 
discourse representation and discuss the 
impact of refined discourse structure on 
question answering.   
1 Introduction 
In a real-world setting, questions are not asked in 
isolation, but rather in a cohesive manner that 
involves a sequence of related questions to meet 
user’s information needs. The capability to interpret 
and answer questions based on context is important. 
For example, Figure 1 shows an example of a series 
of context questions. In this example, the 
interpretation of Q2 and Q4 depends on the 
resolution of “it” and “this” from the context 
respectively. Although neither Q3 nor Q6 requires 
any anaphora resolution, the interpretation of Q3 
depends on Q2 while the interpretation of Q6 
depends solely on itself. Furthermore, in Q5, there 
are no explicit references. Its interpretation depends 
on a preceding question (e.g.,Q4), however, in a 
different manner.     
This example indicates that interpreting each of 
these questions and extracting answers needs to be 
situated in a particular context as the QA session 
proceeds. There are situations where a question is 
“complete” enough and its interpretation does not 
depend on the previous questions (Q6).  There are 
also situations where the interpretation of a question 
depends on preceding questions no matter whether it 
requires anaphora or ellipsis resolution. Based on 
these observations, a natural question to ask is what 
makes the use of discourse differently in different 
situations? What is the role of discourse in context 
question answering? 
To address these questions, a key issue, in our 
mind, is that every question and its answer have a 
discourse status with respect to an entire QA session. 
This discourse status includes two aspects. The first 
aspect relates to discourse roles of entities in a 
question and the corresponding answer. Entities (such 
as noun phrase, verb phrase, preposition phrase, etc) 
in a question carry distinctive roles that indicate what 
is the topic or focus of a question in terms of the 
overall information seeking discourse. Topic relates 
to the “aboutness” of a question and focus relates to a 
specific perspective of the topic. The second aspect 
of discourse status relates to discourse transitions that 
indicate how discourse roles are changed from one 
question to another as the interaction proceeds and 
how such changes reflect the progress of user 
information needs. Both discourse roles and 
discourse transitions determine whether the context is 
useful, and if so, how to use the context to interpret a 
question. 
This paper takes an initial attempt to investigate 
the discourse status for context question answering. 
In particular, it motivates a semantic-rich discourse 
representation that captures both discourse roles of a 
question and discourse transitions between questions. 
Through examples, this paper further discusses the 
potential impact of this refined discourse structure on 
context question answering.   
Q1: What is the name of the volcano that destroyed 
the ancient city of Pompeii?  
Q2: When did it happen?  
Q3: how many people were killed?  
Q4: how tall was this volcano? 
Q5: Any pictures?  
Q6: Where is Mount Rainier?  
 
Figure 1: An example of context questions 
2 Semantic-rich Discourse Modeling 
For processing single questions, an earlier study 
shows that an impressive improvement can be 
achieved when more knowledge-intensive NLP 
techniques are applied at both question and answer 
processing level (Harabagiu et al., 2000). For context 
questions, a parallel question would be whether rich 
contextual knowledge will help interpret subsequent 
questions and extracting answers. To address this 
question, we propose a semantic-rich discourse 
modeling that captures both discourse roles of 
questions and discourse transitions between 
questions, and investigate its usefulness in context 
question answering.  
2.1 Discourse Roles  
In context question answering, each question is 
situated in a context. In addition to the semantic 
information carried by important syntactic entities 
(such as noun phrase, verb phrase, preposition 
phrase, etc), each question also carries distinctive 
discourse roles with respect to the whole question 
answering discourse. Specifically, the discourse roles 
can be categorized based on both the informational 
and intentional perspectives of discourse (Hobbs, 
1996), as well as the presentation aspect of both 
questions and answers.   
The intentional perspective relates to the purpose 
of a question. In a fully interactive question 
answering environment, instead of asking questions, 
a user may need to reply to a clarification question 
prompted by the system or may need to simply ask 
for a confirmation. Therefore, it is important to 
capture the intention from the user (Grosz and Sidner 
1986). The informational perspective relates to the 
information content of a question, in particular, the 
topic and the focus based on the semantics of the 
content. In addition to the intentional and 
informational aspects, there is also a presentational 
aspect of discourse that relates to both the input 
modality (i.e., questions) and the output modality 
(i.e., answers). For example, a user may explicitly ask 
for images or pictures of a person or event. The 
presentation aspect is particularly important to 
facilitate multimodal multimedia question answering. 
Therefore, for a given question, three types of 
discourse roles: Intent, Content, and Media can be 
captured to reflect the intentional, informational, and 
presentational perspectives of discourse respectively.   
These discourse roles can be further characterized 
by a set of features. For example, Intent can be 
represented by Act and Motivator, where Act indicates 
whether the user is requesting information from the 
system or replying to a system question. Motivator 
corresponds to the information goal as to what type 
of action is expected from the system, for example, 
whether information retrieval or confirmation (Chai 
et al, 2003). We will not elaborate Intent here since it 
has been widely modeled for most dialog systems.  
 Content can be characterized as Target, Topic and 
Focus. Target indicates the expected answer type such 
as whether it should be a proposition (e.g., for why 
and how questions), or a specific type of entity (e.g., 
TIME and PLACE).  
Topic indicates the “aboutness” or the scope 
related to a question. Focus indicates the current 
focus of attention given a particular topic. Focus 
always refers to a particular aspect of Topic. Since the 
informational perspective of discourse should capture 
the semantics of what has been conveyed, Topic and 
Focus are linked with the semantic information of a 
question, for example, semantic roles as described in 
(Gildea and Jurafsky 2002). Semantic roles concern 
with the roles of constitutes in a question in terms of 
its predicate-argument structure. The discourse roles 
link the semantic roles of individual questions 
together with respect to the discourse progress 
through Topic and Focus. 
For example, Topic can be of type Activity or Entity. 
Activity can be further categorized by ActType, 
Participant, and Peripheral.  ActType indicates the type 
of the activity; Participant indicates entities that are 
participating in the activity with different semantic 
roles. Peripheral captures auxiliary information such 
 
Content
Target: $NAME 
Topic: Activity
ActType: Destroy [Term: “destroy”]
Participant1: Entity
SemRole: Agent
SemType: volcano
Id: ?
Term: “the volcano” 
Participant2: Entity
SemRole: Theme
SemType: city
Id: “Pompeii”
Term: “Pompeii”        
Focus: Topic.Activity-Participant1
[Element: Name 
[Value: ?; Term: “name” ]]
Intent
Act: Request
Motivator: AnsRequest
Media
Format: Text
Genre: Default
 
Figure 2: Discourse roles for Q1.   
as the time, place, purpose, and reason for such an 
activity. Entity can be categorized by SemRole, 
SemType, Id, Element, and Constraint. SemRole 
indicates the semantic role of the entity in a particular 
activity (if any). SemType represents the semantic 
type of the entity. Element indicates the specific 
features associated with the entity. Constraint 
specifies the constraints need to be satisfied to 
identify the entity, and Id specifies the particular 
identifier of the entity that particularly corresponds to 
pronouns, demonstratives, and definite noun phrases. 
Media indicates the desired information media, 
which can be further characterized as Format and 
Genre as shown in Figure 2. Format indicates whether 
it is an image, a table, or text, etc. Genre specifies the 
answer needs such as summary or list. If it is a list, 
how many should be in the list as in the question 
“number ten largest cities in the world.” 
Figure 2 shows the representation of discourse 
roles of Q1 using typed feature structures (Carpenter 
1992), where Intent indicates that the user is 
requesting for the system to retrieve an answer. Topic 
indicates the topic of Q1 is a Destroy Activity, which 
has two participants. The first participant is some 
kind of unknown volcano that takes the role of Agent 
in the activity (i.e., the destroyer). The second 
participant is the city of Pompeii that takes the role of 
Theme indicating the thing destroyed. The Focus of 
Q1 is about the name (i.e., Element) of the entity in 
the first participant (i.e., Participant1) in the Topic 
representation.  
The granularity of discourse roles can be varied. 
The finer the granularity, the better is the use of 
context for inference (as discussed later). However, 
the finer granularity also implies deeper semantic 
processing. This semantic rich representation can be 
used to generate other representations such as queries 
based on weighted terms for information retrieval or 
logical forms for deduction and inference (Waldinger 
et al., 2003). Furthermore, this representation is able 
to keep all QA sessions in a structured way to support 
inference, summarization, and collaborative fusion as 
described later.   
2.2 Discourse Transitions 
Transitions from one question to another also 
determine how context will be used in interpreting 
questions and retrieving answers. In this section, we 
use query formation as an example to illustrate the 
role of different types of discourse transitions.  
Discourse transitions also correspond to the 
intentional, informational, and presentational 
perspectives of discourse. Intentional transitions are 
closely related to Grosz and Sidner’s “dominance” 
and “satisfaction precedence” relations, which are 
more relevant to plan-based discourse (Grosz and 
Sidner, 1986). Here we focus on informational 
transitions and presentational transitions that are 
more relevant to QA systems since they are targeted 
for information exchange.   
Informational transitions are mainly centered 
around Topics of questions. In context question 
answering, how questions are related to each other 
depends on how “topics” of those questions evolve. 
Currently, we categorize information transitions into 
three types: Topic Extension, Topic Exploration, and 
Topic Shift.  
 
Topic Extension 
A question concerns a similar topic as that of a 
previous question, but with different participants, 
peripheral, or constraints. It has the following 
subcategories: 
Constraint Refinement 
A question is about a similar topic with additional or 
revised constraints. For example:  
Q7: What’s the crime rate in Maryland and Virginia? 
Q8: What is it ten years ago?  
For another example:  
Q9: What’s the crime rate in Maryland and Virginia? 
Q10: What was it in Alabama and Florida? 
In both examples, both questions share the topic of 
“crime rate”, but concerning different crime rates 
with different constraints.  Interpreting the second 
question requires not only identifying constraints, but 
also the relations between constraints. In the first 
example, the constraints from Q7 need to be used to 
form a query for Q8. However, constraints from Q9 
should not be used for Q10.  
 
Participant Shift 
A question is about a similar topic with different 
participants.  
For example:  
Q11: In what country did the game of croquet 
originate?  
Q12: What about soccer? 
In this example, both questions are about the 
origination of a certain sport. The Content structure 
for both questions are the same except for the 
Participant role, which in Q11 is “croquet” and in Q12 
is “soccer”. Therefore, the query created for Q12 
would be {country, soccer, originate}, the keyword 
“croquet” should not appear in the query list.   
 
Topic Exploration  
Two questions are concerning the same topic, but 
with different focus (i.e., asking about different 
aspects of the topic).  For example,  
Q13: What is the name of the volcano that destroyed the 
ancient city of Pompeii? 
Q14: When did this happen?  
In this example, “this” in Q14 refers to the same 
activity topic in Q13, but focus on the TIME 
peripheral information about the activity.  
In the following example,  
Q15: Where is Mount Rainier?  
Q16: How tall is it?  
Q15 asks about the location of Mount Rainier (which 
is an entity topic) and Q16 asks about a different 
aspect (i.e., the height) of the same entity topic. In 
both examples, significant terms representing the 
Topic from the preceding question can be merged 
with the significant terms in the current question to 
form a query.   
 
Topic Shift  
Two consecutive questions could ask about two 
different topics.  Different topic shifts indicate 
different semantic relations between two questions.  
Activity Topic shifts to another Activity Topic 
In the following example,  
Q17: What is the name of the volcano that destroyed the 
ancient city of Pompeii? 
Q18: How many people were killed?  
The topic of both questions concerns about certain 
activities. This activity shift indicates that “kill” 
activity is a consequence of “destroy” activity (i.e., 
Q18 is a consequence of Q17).  
Other relations can also be entailed from such a 
transition such as “effect-cause” relation as in the 
following example (Harabagiu et al. 2001):  
Q19: Which museum in Florence was damaged by a 
major bomb explosion in 1993? 
Q20: How much explosive was used?  
 
Activity Topic shifts to Entity Topic 
In the example: 
Q21: What is the name of the volcano that destroyed the 
ancient city of Pompeii? 
Q22: How tall is this volcano?  
The topic of Q21 is an activity of “destroying” and 
the focus is the agent of the activity “the volcano”. 
This focus becomes the topic of Q22. This transition 
indicates a further probing of a particular participant 
in an activity that can be independent of the activity 
itself. Therefore, the terms in Q21 will not be helpful 
in setting up the stage for processing Q22. Q21 
should be used only to resolve reference to the 
definite noun phrase “this volcano”.  
Related to the presentational perspective of a QA 
discourse, we currently only identify: Media Shift.  
This relation indicates that two questions are about 
the same information content, but with different 
preference of media presentation.  
For example,  
Q25: how tall is Mount Vesuvius? 
Q26: Any pictures? 
Q26 is asking for the images of the Mount Vesuvius. 
This indicates that the backend should perform image 
retrieval rather than text retrieval.  
In summary, given two consecutive questions (Q
i
, 
Q
i+1
), a certain transition exists from Q
i
 to Q
i+1
. 
These transitions determine how the context, for 
example, proceeding questions and answers can be 
used in interpreting the following question and 
identifying the potential answers. Here we only list 
several examples to show the importance of these 
transitions, which are by no means complete. We 
plan to identify a list of salient transitions for 
processing context questions as well as their 
implications (e.g., semantic relations) in interpreting 
context questions.  
2.3 Discourse Processing 
Given the above discussion, the goal of discourse 
modeling for context question answering is to 
automatically identify the discourse roles of a 
question and discourse relations between questions as 
the QA session proceeds. This may be a difficult task 
that requires rich knowledge and deep semantic 
processing. However, the recent advancement in 
semantic processing and discourse parsing has 
provided an excellent foundation for this task.  
The discourse roles are higher-level abstracts of 
the semantic roles as those provided in FrameNet 
(Baker et al., 1998) and Propbank (Kingsbury and 
Palmer 2002). Recent corpus-based approaches to 
identify semantic roles (Roth et al 2002, Gildea and 
Jurafsky 2002; Gildea and Palmer 2002; Surdeanu et 
al., 2003) have been successful in identifying domain 
independent semantic relations with respect to the 
predicate-argument structure. Furthermore, recent 
work also provides discourse annotated corpora with 
rhetorical relations (Carlson, et al., 2003) and 
techniques for discourse paring for texts (Soricut and 
Marcu, 2003).  All these recent advances make the 
semantic-rich discourse modeling possible.  
For example, a collection of context questions 
(and answers) can be annotated in terms of their 
discourse roles and relations. Specifically, the 
following information can be either automatically 
identified or manually annotated: 
• Syntactic structures automatically identified from 
a parser (Collins, 1997); 
• Semantic roles of entities in the question (Gildea 
and Jurafsky 2002; Gildea and Palmer 2002; 
Surdeanu et al., 2003); 
• Discourse roles either manually annotated or 
identified by rules that map directly from semantic 
roles to discourse roles.  
• Discourse transitions automatically determined 
once discourse roles are identified for each 
question.  
• Semantic relations between questions manually 
annotated. 
• Answers provided by the system.  
Based on this information, important features can be 
identified. Different learning models such as decision 
trees or Bayesian classifier can be applied to learn the 
classifier for discourse roles and relations. Strategies 
can be built to take into account of discourse roles 
and relations from preceding questions and answers 
to process a subsequent question and extract answers. 
These models can then be applied to process new 
context questions.  
3 Refined Discourse Structure in Context 
Question Answering 
Based on the above discussion, during the question 
answering process, a discourse structure can be 
created to capture the discourse roles of each 
ActType
Activity
City
?
Volcano
Name
Theme
Agent
Entity
Destroy
“Pompeii”
Entity
Id
Id
Participant1
SemType
SemRole
Participant2 SemRole
SemType
Element
Topic
Focus
“?”
    
ActType
Activity
City
“Mount Vesuvius”
Volcano
Name
Theme
Agent
Entity
Destroy
“Pompeii”
Entity
Id
Id
Participant1
SemType
SemRole
Participant2 SemRole
SemType
Element
Topic
“Mount Vesuvius”
Time
?
Peripheral
Value
Focus
 
 
(a) Discourse repreentation after processing Q1                (b) Discourse representation after processing Q2             
 
 
ActType
Activity
City
“Mount Vesuvius”
Volcano
Name
Theme
Agent
Entity
Destroy
“Pompeii”
Entity
Id
Id
Participant1
SemType
SemRole
Participant2 SemRole
SemType
Element
Topic
“Mount Vesuvius”
Time
79 AD
Peripheral1
Value
Focus
Consequence
Activity
EntityKill
Patient
Person
SizeOfSet ?
Peripheral2
Type
ActType
Participant1
SemType
SemRole
Element
Value
Value
        
ActType
Activity
City
“Mount Vesuvius”
Volcano
Name
Theme
Agent
Entity
Destroy
“Pompeii”
Entity
Id
Id
Participant1
SemType
SemRole
Participant2 SemRole
SemType
Element
Topic
“Mount Vesuvius”
Time
79 AD
Peripheral
Value
Focus
Height
Element
Value
?
Activity
EntityKill
Patient
Person
SizeOfSet Thousands
Peripheral2
Type
ActType
Participant1
SemType
SemRole
Element
Value
Consequence
 
 
(c) Discourse representation after processing Q3              (d) Discourse representation after processing Q4             
question and discourse relations between questions.  
Similar to information extraction for free texts, this 
refined discourse structure captures the salient 
information extracted from the question answering 
process. This discourse provides a structured 
information space that indicates what type of 
information has been exchanged and how 
information obtained at different stages is related. In 
other words, we can also consider this representation 
as the “mental map” of user information needs. This 
mental map will potentially provide a basis to 
improve question interpretation and answer 
extraction through inference, summarization, and 
collaborative question answering.    
3.1 Discourse Representation 
The typed feature structures can be represented as 
Directed Acyclic Graph (DAG). Thus the described 
discourse structure can be represented as semantic 
networks using DAGs. For example, Figure 3 shows 
the discourse representation after processing the each 
of the first four questions in Figure 1. In this network, 
each node is either a specific value (i.e., leaf nodes) 
or a typed feature structure itself (i.e., internal node). 
Each directed link corresponds to a particular feature. 
Note that because of the space limit, not everything 
represented in the feature structure in Figure 2 is 
shown here in the semantic network. For example, 
the type of an activity (e.g., Destroy) by itself is a 
feature structure (in Figure 2) that further consists of 
the specific term used in the question.  This term is 
not shown but is included in the semantic nets.  
As context question answering proceeds, the 
semantic network(s) for discourse grows, with 
different pointers of Topic and Focus. For example, 
Figure 3(a) represents Q1, where Topic points an 
Activity feature structure and Focus points to the Name 
Element of the Participant1 in the Activity. From Q1 to 
Q2, there is a transition of Topic Exploration which 
indicates that Q2 is about the same topic, but with a 
different focus. Therefore, in Figure 3(b), the Topic 
points to the same activity, but the Focus now points 
to the peripheral Time information of that activity.  
Next, Q3 is about a different topic involving activity 
Kill. However, since there is a consequence relation 
from Q2 to Q3, the activity asked in Q3 actually 
fulfills the Peripheral role of Consequence for the 
previous activity as shown in Figure 3(c). Finally, in 
Q4, there is a gap between Q3 and Q4, however, 
there is a transition of Probing from Q2 to Q4. Now 
the Topic becomes the Participant in Q1 as shown in 
Figure 3(d).  
3.2 Potential Impacts 
The growth of the semantic networks represents the 
overall information needs of a user and how such 
information needs are related. Since this is a 
structured representation, it can be queried and used 
to facilitate context question answering, for example, 
in the following aspects: 
• Query expansion and answer retrieval 
• Inference and summary for question answering 
• Collaborative question answering 
To process questions, most systems will first form 
a query of keywords to represent the current question 
and to retrieve relevant passages that may contain the 
potential answers. In context question answering, 
since the interpretation of a question may depend on 
preceding questions, some keywords from preceding 
questions may need to be included in the query for 
the current question. The fine-grained discourse 
structure will enhance answer retrieval through more 
controlled selection of terms from preceding 
questions and answers. For example, strategies can be 
developed to select query terms depending on the 
discourse relations. Different discourse roles and 
transitions may lead to different weighting schemes 
for query expansion.  
Furthermore, the information captured in the 
discourse structure can help make predication about 
what the user information need is and therefore 
provide more intelligent services to help user find 
answers. For example, semantic and discourse 
relations between different topics and focuses of a 
series of questions can help a system infer and predict 
the overall interest of a user. Although answers to 
each question may come from different sources, 
based on the structured discourse (e.g., in semantic 
network), the system can aggregate information and 
generate summaries.  
Another potential impact of the refined structured 
discourse is to facilitate collaborative question 
answering. Very often, various users may have a 
similar interest about a set of topics. The structured 
discourse built for one user can be used to help 
answers questions from another user. A user may 
have a certain information goal in mind, but does not 
know what types of questions to ask.  Therefore, a 
user’s question may be very general and vague, such 
as “what happened to Pompeii?” This question needs 
to be decomposed into a set of smaller questions. The 
discourse structure that connects different aspects of 
topics together can provide some insight on how such 
decomposition should be made. Furthermore, the 
discourse structure from a skilled user can enable the 
system to intelligently direct a novice user in his 
information seeking process.  
4 Discussion 
TREC 10 Question Answering Track initiated a 
context-task that was designed to investigate the 
system capability to track context through a series of 
questions. As described in (Voorhees 2001), there 
were two unexpected results of this task. First, the 
ability to identify the correct answer to a question in 
the later series had no correlation with the capability 
to identify correct answers to preceding questions. 
Second, since the first question in a series already 
restricted answers to a small set of documents, the 
performance was determined by whether the system 
could answer a particular type of question, rather than 
the ability to track context. Because of these 
unexpected results, the context task has been stopped 
in the following TREC evaluations (Voorhees 2002).  
The reasons that TREC 10 did not achieve the 
expected results, in our opinion, lie in two aspects. 
The first aspect relates to the uniqueness of open 
domain context question answering. In open domain 
QA, first, there may be many occurrences of correct 
answers in various part(s) of document(s). Second, 
there may be multiple paths (e.g., different 
combination of key query terms) that can lead to one 
occurrence of the correct answer. Therefore, the 
correct answer to a previous question may not be 
critical in finding answers to subsequent questions. 
This phenomenon may provide an opportunity to find 
answers without explicitly modeling context (i.e., by 
keeping track of the discourse objects from answers), 
but rather identifying and using relevant context.  
For example, in the LCC system (that achieved 
the best result for the context task in TREC 10), the 
discourse was not explicitly represented (Harabagiu 
et al 2001). Instead of resolving references using 
discourse information, the LCC system first identifies 
the question that contains the potential referents and 
uses those questions and the current question to 
identify the target paragraph. Thus, question 
interpretation does not depend on the answers, but 
rather depends on the context that is dynamically 
identified as a list of preceding questions. Now the 
question is whether the system will achieve even 
better results (e.g., correctly find answers to the rest 
of the eight questions) with some context 
representation?  
Another more important question to be asked is 
whether the design of the context task just happened 
to provide an opportunity to achieve good results 
without modeling the context. As discussed in 
(Harabagiu et al., 2001), answers to 85% of context 
questions actually occurred within the same 
paragraph as the answers to the previous questions. 
Therefore, just using preceding questions, the system 
was able to find the target paragraph and the final 
answer really depended on the capability to identify 
different types of answers in that paragraph. What if 
a series of questions were designed differently so that 
questions are related but answers are scattered in 
different documents or paragraphs.  Will the shallow 
processing of discourse succeed in finding the 
answers?  
Furthermore, the ultimate goal of QA systems is 
to be able to access information from different 
sources (e.g., unstructured text or structured 
database) and to provide intelligent dialog capability.  
One important question we need to address is what 
kind of discourse representation will be sufficient to 
support these capabilities? For example, to access 
structured databases, the answer to a previous 
question usually narrows down the search spaces in 
the database for subsequent questions. Thus, previous 
answers usually determine where in the database an 
answer can be found. Therefore, it is important to 
keep track of previous questions and answers in some 
kind of structure for later use.  
The second reason that TREC 10 did not achieve 
expected results relates to evaluation methodology. In 
context QA, good performance depends on two 
important components: the capability of representing 
and using the relevant context (both explicitly or 
implicitly) and the general capability of interpreting 
questions and extracting answers. The level of 
sophistication in either component will influence the 
final performance. Thus, by comparing the final 
answers to each context question, the evaluation of 
the context task in TREC 10 was not able to isolate 
the effect of one component from another. It is not 
feasible to identify that when an answer is not 
identified, whether it is because of poor 
representation and use of the discourse information or 
it is because of the general limitations of the 
capability to process certain types of questions. 
Therefore, to study the role of discourse in context 
question answering, a more controlled evaluation 
mechanism is desired. For example, one approach is 
to keep the general processing capability as a 
constant and vary the representations of discourse 
and strategies to use the discourse so that their 
different impacts on the final answer extraction can 
be learned.  
As a summary, the experience in the TREC 10 
context task is very valuable. It does not discount the 
importance of context modeling for context 
questions. But rather, it motivates a more in-depth 
investigation of the role of discourse in context 
question answering.  
5 Conclusion 
Questions are not asked in isolation, but rather in a 
cohesive manner that involves a sequence of related 
questions to meet user’s information needs. It is 
important to understand the role of discourse to 
support this cohesive question answering.  
By all means, a QA discourse can be represented 
as coarse as a list of keywords extracted from 
previous questions or as sophisticated as a fine-
grained representation as described in this paper. 
There is a balance between how much we like to 
represent the context and how far we can get there.  
Given recent advances in text-based domain 
independent semantic processing and discourse 
parsing, as well as the availability of rich semantic 
knowledge sources, we believe it is the time to start 
from the other end of the spectrum to examine the 
possibility and impact of semantic-rich discourse 
representation for open-domain question answering.     

References 
Baker, C., Fillmore, C., and Lowe, J. 1998. The Berkeley 
FrameNet project. In Proceedings of COLING/ACL, pp. 
86-90, Montreal, Canada 
Carpenter, R.1992. The Logic of Typed Feature Structures. 
Cambridge University Press.  
Chai, J., Pan, S., and Zhou, M. 2003. MIND: A Context-
based Multimodal Interpretation Framework in Conver-
sational Systems, Natural, Intelligent and Effective In-
teraction in Multimodal Dialogue Systems,  Eds. O. 
Bernsen ,  L. Dybkjaer and  J.  van Kuppevelt, Kluwer 
Academic Publishers. 
Carlson, L., Marcu, D., and Okurowski, M. 2003. Building 
a discourse-tagged corpus in the framework of Rhetori-
cal Structure Theory. In Jan van Kuppevelt and Ronnie 
Smith, editors, Current Directions in Discourse and 
Dialogue, Kluwer Academic publishers.  
Collins, M. 1997. Three Generative, Lexicalized Models 
for Statistical Parsing. In Proceedings of the 35th Annual 
meeting of the Association for Computational Linguis-
tics (ACL 1997): 16-23, Madrid, Spain.  
Fillmore, C. and Atkins, B. 1998. FrameNet and lexico-
graphic relevance, Proceedings of the First Interna-
tional Conference on Language Resources and 
Evaluation, Granada, Spain. 
Grosz, B. J. and Sidner, C. 1986. Attention, intention, and 
the structure of discourse. Computational Linguistics, 
12(3):175-204. 1986. 
Gildea, D. and Jurafsky, D. 2002. Automatic Labeling of 
Semantic Roles. Computational Linguistics, 28(3):245-
288.  
Gildea, D. and Palmer, M. 2002. The Necessity of Parsing 
for Predicate Argument Recgnition. In Proceedings of 
the 40th Meeting of the Association for Computational 
Linguistics (ACL 2002): 239-246, Philadelphia, PA.  
Harabagiu, S., Pasca, M., and Maiorano, S. Experiments 
with Open-domain Textual Question Answering. In 
Proceedings of the 18th International Conference on 
Computational Linguistics (COLING-2000), 2000 
Harabagiu, S., et al., Answering Complex, List and Con-
text Questions with LCC’s Question-Answering Server. 
In the Proceedings of TREC 2001, 2001. 
Hobbs, J. 1996. On the relations between the informational 
and Intentional Perspectives on Discourse. In Burning 
Issues in Discourse: An inter-Disciplinary Account  
(eds. E. Hovy and D. Scott), volume 151 of NATO ASI 
Series, Series F: Computer and Systems Sciences, pp 
139-157. Springer-Verlag, Berlin, Germany 
Kingsbury, P. and Palmer, M. 2002. From Treebank to 
Propbank. In Proceedings of the 3rd International Con-
ference on Language Resources and Evaluation (LREC-
2002), Las Palmas, Canary Islands, Spain.  
Roth, D., et al.  2002. Question Answering via Enhanced 
Understanding of Questions. Proceedings of 
TREC2002.  
Surdeanu, M., Harabagiu, S., Williams, J., and Aarseth, P. 
2003. Using Predicate-Argument Structures for Infor-
mation Extraction. In Proceedings of the 41th Meeting of 
the Association for Computational Linguistics (ACL 
2003), Sapporo, Japan.  
Soricut, R. and Marcu, D. 2003. Sentence Level Discourse 
Parsing using Syntactic and Lexical Information. In 
Proceedings of HLT-NAACL. Edmonton, Canada. 
Waldinger, R., et al., 2003. Deductive Question Answering 
from Multiple Resources, New Directions in Question 
Answering, AAAI, 2003 
Voorhees, E. 2001. Overview of TREC 2001 Question 
Answering Track. Proceedings of TREC2001.  
Voorhees, E. 2002. Overview of TREC 2002 Question 
Answering Track. Proceedings of TREC2002.  
