Content-based Document Classification with Highly Compressed Input Data

@inproceedings{Merkl1995ContentbasedDC,
  title={Content-based Document Classification with Highly Compressed Input Data},
  author={Dieter Merkl},
  year={1995}
}
One of the major obstacles for the application of artificial neural networks to real-world problems is the rather time-consuming task of training. In this paper we will demonstrate that considerable acceleration with equal classification results may be achieved by the utilization of highly compressed input data for a self-organizing map. As the basis for the experiments we use an application in the area of software reuse, namely the structuring of software components. As a result we are able to… CONTINUE READING
18 Citations
9 References
Similar Papers

References

Publications referenced by this paper.
Showing 1-9 of 9 references

The Effects of Lateral Inhibition on Learning Speed and Precision of a SelfOrganizing Feature Map

  • Adelaide
  • Proc Australian Conf on Neural Networks
  • 1995

Application of Self-Organizing Feature Maps with Lateral Inhibition to Structure a Library of Reusable Software Components

  • D. Merkl, Tjoa, M A, G. Kappel
  • Proc IEEE Int ’ l Conf on Neural Networks
  • 1994
1 Excerpt

1992).A User’s Guide to PlaNet Environment for Running, and Looking into a PDP Network

  • Y. Miyata
  • Version 5.8. School of Computer and Cognitive…
  • 1992
1 Excerpt

A Comparison of Text Retrieval Models

  • R TurtleH., W. B. Croft
  • The Computer Journal
  • 1992

Similar Papers

Loading similar papers…