Skip to main content

2020 | OriginalPaper | Buchkapitel

23. An Introduction to Neural Networks

verfasst von : Prof. Max Bramer

Erschienen in: Principles of Data Mining

Verlag: Springer London

Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.

search-config
loading …

Abstract

This chapter gives an introduction to the important topic of neural networks, computing systems based loosely on the connections between the neurons in the brain, which are increasingly widely used in data mining as well as other areas. A feed-forward neural network with backpropagation is described and its operation explained in detail. It is shown how such a network can be used for classification tasks and experimental results with two well-known datasets are presented. The chapter ends with some cautionary remarks about the drawbacks involved in using a neural net, relating mainly to inscrutability and the risk of inadvertently building bias into the system’s conclusions.

Sie haben noch keine Lizenz? Dann Informieren Sie sich jetzt über unsere Produkte:

Springer Professional "Wirtschaft+Technik"

Online-Abonnement

Mit Springer Professional "Wirtschaft+Technik" erhalten Sie Zugriff auf:

  • über 102.000 Bücher
  • über 537 Zeitschriften

aus folgenden Fachgebieten:

  • Automobil + Motoren
  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Elektrotechnik + Elektronik
  • Energie + Nachhaltigkeit
  • Finance + Banking
  • Management + Führung
  • Marketing + Vertrieb
  • Maschinenbau + Werkstoffe
  • Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

Springer Professional "Technik"

Online-Abonnement

Mit Springer Professional "Technik" erhalten Sie Zugriff auf:

  • über 67.000 Bücher
  • über 390 Zeitschriften

aus folgenden Fachgebieten:

  • Automobil + Motoren
  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Elektrotechnik + Elektronik
  • Energie + Nachhaltigkeit
  • Maschinenbau + Werkstoffe




 

Jetzt Wissensvorsprung sichern!

Springer Professional "Wirtschaft"

Online-Abonnement

Mit Springer Professional "Wirtschaft" erhalten Sie Zugriff auf:

  • über 67.000 Bücher
  • über 340 Zeitschriften

aus folgenden Fachgebieten:

  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Finance + Banking
  • Management + Führung
  • Marketing + Vertrieb
  • Versicherung + Risiko




Jetzt Wissensvorsprung sichern!

Fußnoten
1
To avoid possible confusion, there is no reason why all the layers should have the same number of nodes. Also, a layer with just a single node is permitted.
 
2
In the neural network literature, the links between nodes are often described as the connections between neurons.
 
3
The training data used with Neural Networks can have any number of output values, not just the single value (the class value) included in the training data used for classification tasks elsewhere in this book. The use of a neural net for classification is discussed in Section 23.6.
 
4
As a variation, a different bias weight can be associated with each hidden and output node. In this book we will assume that every node in a layer always has the same value of the bias weight.
 
5
Numbers are generally displayed to four decimal places but full accuracy is retained in future calculations.
 
6
A factor of 0.5 is included to make some of the mathematical calculations at a later stage easier. It makes very little practical difference.
 
7
Differential calculus is a branch of mathematics concerned with the rate of change of one variable with respect to another. For example, velocity is the rate of change of distance from a fixed point ‘with respect to time’, i.e. as time changes. Acceleration is the rate of change of velocity with respect to time.
 
8
The effect of this approach is that if the gradient is positive, the weight is reduced; if it is negative, the weight is increased.
 
Metadaten
Titel
An Introduction to Neural Networks
verfasst von
Prof. Max Bramer
Copyright-Jahr
2020
Verlag
Springer London
DOI
https://doi.org/10.1007/978-1-4471-7493-6_23

Neuer Inhalt