Since I’m currently studying for my next oral diploma exam on “introduction to neural information processing” (called Neuroinformatics and Data Mining) there are, of course, some Internet references of interest (mainly German):
- lecturer’s / examiner’s slides (DE)
- Scriptum on Grundstudium.info (DE)
- David Kiesel’s Book (DE)
- NeuronalesNetz.de, an overview (DE)
- Demo I of Kohonen Self-Organizing Feature Map/Self-Organising Maps or Demo II (ball-catching cart, DE)
- GNGDemo on different algorithms, e.g. SOM, Hard Competitive
- K-Means vs. Fuzzy K-Means Demo. It also has a demo on dendrogram demonstrated deploying with HAC (Hierarchical Agglomerative Clustering) algorithms such as single-linkage.
- DemoGNG, implementing diverse learning methods
- 2 more demos for K-Means and Fuzzy K-Means including very good explanation for each algorithm (links at the top)
- diverse articles at wikipedia (German and English) varying from overviews on ANN/KNN and Data Mining (DE, EN) to detailed ones, e.g. on RBF Networks.
- some advanced reading and examples at Peltaron, developer of synaptic.
- Tutorial on PCA, examples incl.
Books I recommend:
A Classic: Raúl Rojas: “Theorie der neuronalen Netze.” First print published 1993 at Springer-Verlag, Berlin. ISBN: 3540563539 or it’s english version “Neural Networks” (Rojas says: “The English version is almost a new book”) ISBN: 3540605053. I only read the german version, so far. This I can recommend as it’s language is easily understandable which helps in the learning process. I gives a good overview over established neural modells without loosing details. It’s more on the theoretic side, though.
Detlef Nauck et all.: “Neuronale Netze und Fuzzy-Systeme.” Published at Vieweg 1993. ISBN: 3528052651. Even though it reads Fuzzy-Systems in the title about the first half of the book introduces artificial and biological neural methods in a quite practical manner. It gives many easy and basic examples helping to understand the principles, pros and cons of the many approaches.