Big Data

136 Big Data: Techniques and Technologies in Geoinformatics

7.1.1.2 Unsupervised Learning

Unsupervised learning, on the other hand, is primarily concerned with clustering

or grouping of examples and can be used effectively for anomaly detection and

is also an effective preprocessing step as in the case of dimensionality reduction.

The main purpose of unsupervised learning is to nd some structures to unla-

belled training examples. Once a structure is found, it can be exploited in a num-

ber of ways. First, it can be used to aggregate the data in some way arriving at a

subset of representative or generalized examples. Second, any new examples that ...