<?xml version="1.0" standalone="yes"?>
<Paper uid="H05-2004">
  <Title>Demonstrating an Interactive Semantic Role Labeling System</Title>
  <Section position="3" start_page="6" end_page="6" type="metho">
    <SectionTitle>
2 The SRL System Architecture
</SectionTitle>
    <Paragraph position="0"> Our system begins preprocessing raw text by using sentence segmentation tools (available at http://l2r.cs.uiuc.edu/[?]cogcomp/tools.php). Next, sentences are analyzed by a state-of-the-art syntactic parser (Charniak, 2000) the output of which provides useful information for the main SRL module.</Paragraph>
    <Paragraph position="1"> The main SRL module consists of four stages: pruning, argument identification, argument classification, and inference. The following is the overview of these four stages. Details of them can be found in (Koomen et al., 2005).</Paragraph>
    <Paragraph position="2"> Pruning The goal of pruning is to filter out unlikely argument candidates using simple heuristic rules. Only the constituents in the parse tree are considered as argument candidates. In addition, our system exploits a heuristic modified from that introduced by (Xue and Palmer, 2004) to filter out very unlikely constituents.</Paragraph>
    <Paragraph position="3"> Argument Identification The argument identification stage uses binary classification to identify whether a candidate is an argument or not. We train and apply the binary classifiers on the constituents supplied by the pruning stage.</Paragraph>
    <Paragraph position="4"> Argument Classification This stage assigns the final argument labels to the argument candidates supplied from the previous stage. A multi-class classifier is trained to classify the types of the arguments supplied by the argument identification stage.</Paragraph>
    <Paragraph position="5"> Inference The purpose of this stage is to incorporate some prior linguistic and structural knowledge, such as &amp;quot;arguments do not overlap&amp;quot; and &amp;quot;each verb takes at most one argument of each type.&amp;quot; This knowledge is used to resolve any inconsistencies in argument classification in order to generate legitimate final predictions. The process is formulated as an integer linear programming problem that takes as input confidence values for each argument type supplied by the argument classifier for each constituent, and outputs the optimal solution subject to the constraints that encode the domain knowledge.</Paragraph>
    <Paragraph position="6"> The system in this demonstration, however, differs from its original version in several aspects.</Paragraph>
    <Paragraph position="7"> First, all syntactic information is extracted from the output of the full parser, where the original version used different information obtained from different processors. Second, the named-entity information is discarded. Finally, no combination of different parse tree outputs is performed. These alterations aim to enhance the efficiency of the system while maintaining strong performance.</Paragraph>
    <Paragraph position="8"> Currently the system runs at the average speed of 1.25 seconds/predicate. Its performance is 77.88 and 65.87 F1-score on WSJ and Brown test sets (Carreras and M`arquez, 2005) while the original system achieves 77.11 and 65.6 on the same test sets without the combination of multiple parser outputs and 79.44 and 67.75 with the combination.</Paragraph>
  </Section>
  <Section position="4" start_page="6" end_page="6" type="metho">
    <SectionTitle>
3 Goal of Demonstration
</SectionTitle>
    <Paragraph position="0"> The goal of the demonstration is to present the system's ability to perform the SRL task on raw text in real time. An interactive interface allows users to input free form text and to receive the SRL analysis from our system. This demonstration can be found at http://l2r.cs.uiuc.edu/[?]cogcomp/srl-demo.php.</Paragraph>
  </Section>
class="xml-element"></Paper>