<?xml version="1.0" standalone="yes"?>
<Paper uid="W03-2002">
  <Title>Intelligent patent analysis through the use of a neural network: experiment of multi-viewpoint analysis with the MultiSOM model</Title>
  <Section position="7" start_page="4" end_page="4" type="concl">
    <SectionTitle>
6. Conclusion
</SectionTitle>
    <Paragraph position="0"> We have presented a new self-organizing multi-map system. We proposed it as a visualization-based system for scientific and technical information analysis, like patents analysis. The model that this multi-map environment provides is certainly not the map but in its original extended version of intercommunication between multiples maps. Each map representing a particular viewpoint extracted from the data. These viewpoints are related either by the problem to be solved, or by the intercommunication mechanism between the maps.</Paragraph>
    <Paragraph position="1"> We have exposed both the map generation and their intercommunication mechanism. We finally showed how one can evaluate such a viewpoint-oriented approach by comparing it to a global classification approach.</Paragraph>
    <Paragraph position="2"> The advantages of the MultiSOM method seem obvious both in terms of objective evaluation, like the one we proposed, and for the domain experts: the original multiple viewpoints classification approach of MultiSOM tends to reduce the noise which is inevitably generated in an overall classification approach while increasing the flexibility and the granularity of the analyses.</Paragraph>
    <Paragraph position="3"> Moreover, with a global classification method, even if this latter manages overlapping classes, important relationships between some subtopics are hidden in the class profiles and therefore very difficult to precisely characterize.</Paragraph>
    <Paragraph position="4"> Our experiment has also highlighted that our quality evaluation factors that we have proposed can be benefitely used for optimizing the classifications in terms of number of classes, either these classifications are global or they are viewpoint-oriented. This optimization seems to be mandatory when one want to classify documents issued from the Web, where sparseness could usually be a blocking factor.</Paragraph>
  </Section>
class="xml-element"></Paper>