<?xml version="1.0" standalone="yes"?>
<Paper uid="W98-1224">
  <Title>Do Not Forget: Full Memory in Memory-Based Learning of Word Pronunciation *</Title>
  <Section position="6" start_page="0" end_page="0" type="evalu">
    <SectionTitle>
5 Results
</SectionTitle>
    <Paragraph position="0"> Figure 1 displays the generalisatiou acc~acies in terms of incorrectly classified test instances obtained with all performed experiments. The leftmost point in the Figure, f~om which all lines originate, indicates the performance of IBIdegIG when trained on the full data set of 222,601 types, viz. 6.42% incorrectly classified test instances (when computed in terms of incorrectly pronounced test words, IBI-IG pronounces 64.61 of all test words flawlessly).</Paragraph>
    <Paragraph position="1"> The line graph representing the fou~ expemnents in which instance types are removed randomly can be seen as the baseline graph. It can be expected  (fa). Each ranked nearest neighbour is identified by its match (o) or mismatch (x) with the target instance the ranking is computed for, and a number denoting its distance to the target instance. that removing instances randomly leads to a degradation of generalisation performance. The upward curve of the line graph denoting the experiments with random selection indeed shows degrading performanee with increasing numbers of left-out instance types. The relative decrease in generalisation accuracy is 2.0% when 1% of the training material is removed randomly, 3.8% with 2% random removal, 10.7% with 5% random removal, and 20.7% with 10% random removal.</Paragraph>
    <Paragraph position="2"> Surprisingly, the only experiments showing lower performance degradation than removal by random selection are those with class-prediction strength; the other criteria for removing exceptional instances lead to worse degradations. It does not matter whether instance types are removed on grounds of their typicality: apparently, a markedly low, neutral, or high typicality value indicates that the instance type is (on average) important, rather than removable. The same applies to friendly-neighbourhood size: instances with small neighbourhood sizes appear to contribute significantly to performance on test material. It is remarkable that the largest errors with 1% and 2% removal are obtained with the friendly-neighbourhood size criterion: it appears that on average, the instances with few or no nearest neighbours are important in the classification of test material.</Paragraph>
    <Paragraph position="3"> When using class-prediction strength as removal criterion, performance does not degrade until about 5% of the instance types with the lowest strength are removed from memory. The reason is that c|_~ssprediction strength is the only criterion that detects minority ambiguities, i.e., instance types with prediction strength 0.0, that cannot contribute to classification since they are always overshadowed by their counterpart instance types with majority classes, even for their own classification. In the tralni~g set, 9,443 instance types are minority ambiguities, i.e., 4.2% of the instance types (accounting for 3.8% of the instance tokens in the original token set).</Paragraph>
    <Paragraph position="4"> Thus, among the tested methods for reducing the memory needed for storing an instance base in memory-based learning, only two relatively trivial methods are performance-preserving while accounting for a substantial reduction in the amount of memory needed by IB 1-IG: 1. Replacing instance tokens by instance types accounts for a reduction of about 63% of memory needed to store instances, excluding the memory needed to store frequency information.</Paragraph>
    <Paragraph position="5"> When frequency information is stored in two bytes per instance type, the memory reduction is about 54%.</Paragraph>
    <Paragraph position="6"> . Removing instance types that are minority ambigulties on top of the type/token-reduction accounts only for an additional memory reduction of 2%, i.e., for a total memory reduction of 65%; 56% with two-byte frequency information stored per instance.</Paragraph>
  </Section>
class="xml-element"></Paper>