General
Feature extraction involves simplifying the amount of resources required to describe a large set of data accurately. When performing analysis of complex data one of the major problems stems from the number of variables involved. Analysis with a large number of variables generally requires a large amount of memory and computation power or a classification algorithm which overfits the training sample and generalizes poorly to new samples. Feature extraction is a general term for methods of constructing combinations of the variables to get around these problems while still describing the data with sufficient accuracy.
Best results are achieved when an expert constructs a set of application-dependent features. Nevertheless, if no such expert knowledge is available general dimensionality reduction techniques may help. These include:
- Principal component analysis
- Semidefinite embedding
- Multifactor dimensionality reduction
- Multilinear subspace learning
- Nonlinear dimensionality reduction
- Isomap
- Kernel PCA
- Multilinear PCA
- Latent semantic analysis
- Partial least squares
- Independent component analysis
- Autoencoder
Read more about this topic: Feature Extraction
Famous quotes containing the word general:
“There is in general good reason to suppose that in several respects the gods could all benefit from instruction by us human beings. We humans are—more humane.”
—Friedrich Nietzsche (1844–1900)
“A bill of rights is what the people are entitled to against every government on earth, general or particular, and what no just government should refuse, or rest on inference.”
—Thomas Jefferson (1743–1826)
“It has been the struggle between privileged men who have managed to get hold of the levers of power and the people in general with their vague and changing aspirations for equality, for justice, for some kind of gentler brotherhood and peace, which has kept that balance of forces we call our system of government in equilibrium.”
—John Dos Passos (1896–1970)