<?xml version="1.0" standalone="yes"?>
<Paper uid="P98-2201">
  <Title>A Connectionist Approach to Prepositional Phrase Attachment for Real World Texts</Title>
  <Section position="6" start_page="1234" end_page="1234" type="evalu">
    <SectionTitle>
4 Encoding and Network
</SectionTitle>
    <Paragraph position="0"> Architecture.</Paragraph>
    <Paragraph position="1"> Semantic classes were extracted from Wordnet 1.5. In order to encode each word we did not use Word-Net directly, but constructed a new hierarchy (a sub-set of WordNet) including only the classes that corresponded to the words that belonged to the training and test sets. We counted the number of times the different semantic classes appear in the training and test sets. The hierarchy was pruned taking these statistics into account. Given a threshold h, classes which appear less than h% were not included. In this way we avoided having an excessive number of classes in the definition of each word which may have been insufficiently trained due to a lack of examples in the training set. We call the new hierarchy obtained after the cut WordNei'. Due to the large number of verb hierarchies, we made each verb lexicographical file into a tree by adding a root node corresponding to the file name. According to Miller et al. (1993), verb synsets are divided into 15 lexicographical files on the basis of semantic criteria. Each root node of a verb hierarchy belongs to only one lexicographical file. We made each old root node hang from a new root node, the label of which was the name of its lexicographical file. In addition, we codified the name of the lexicographical file of the verb itself.</Paragraph>
    <Paragraph position="2"> There are essentially two alternative procedures for using class information. The first one consists of the simultaneous presentation of all the classes of all the senses of all the words in the 4-tuple. The input was divided into four slots representing the verb, nl, prep, and n2 respectively. In slots nl and n2, each sense of the corresponding noun was encoded using all the classes within the IS-A branch of the WordNet'hierarchy, from the corresponding hierarchy root node to its bottom-most node. In the verb slot, the verb was encoded using the IS_A_WAY_OF branches. There was a unit in the input for each node of the WordNet subset. This unit was on if it represented a semantic class to which one of the senses of the word to be encoded belonged. As for the output, there were only two units representing whether the PP attached to the verb or not.</Paragraph>
    <Paragraph position="3"> The second procedure consists of presenting all the classes of each sense of each word serially. However, the parallel procedure have the advantage that the network can detect which classes are related with which ones in the same slot and between slots. We observed this advantage in preliminary studies.</Paragraph>
    <Paragraph position="4"> Feedforward networks with one hidden layer and  aAccuracy obtained by Brill and Resnik (94) using Resnik's method on a larger test. bThis accuracy is based on 66% coverage.</Paragraph>
    <Paragraph position="5"> a full interconnectivity between layers were used in all the experiments. The networks were trained with backpropagation learning algorithm. The activation function was the logistic function. The number of hidden units ranged from 70 to 150. This network was used for solving our classification problem: attached to noun or attached to verb. The output activation of this network represented the bayesian posterior probability that the PP of the encoded sentence attaches to the verb or not (Richard and Lippmann (1991)).</Paragraph>
  </Section>
class="xml-element"></Paper>