This preview shows page 1. Sign up to view the full content.
Unformatted text preview: 2. Correlation analysis. It studies the correspondence of variables to each other,
such as the x2.
Cluster analysis. It finds groups from a set of objects based on distance
measure. 4. Bayesian network. It represents the causal relationships (what object/event is
caused by which other objects/events) among the variables in the form of a
directed graph, which is computed by using the Bayesian probability theorem.
Like statistical methods, machine learning methods search for the model that is
closest (best matches) the data being tested. However, unlike statistical methods,
the searching space is a cognitive space of n attributes instead of a vector space of
n dimensions. Besides that, most machine learning methods use heuristics in the
Some of the commonly used machine learning methods used in data mining are:
1. Decision tree. Decision trees are a way of representing a series of rules for the
classification of a dataset and consist of nodes and branches. They are built using a
training set of data and are used to classify objects in a dataset. An object's class is
determined by the use of a decision tree by following the path from the root to a
leaf node. It chooses the branches according to the attribute values of the object. A
simple decision tree that may be used for classifying the employees of an
organization into groups based on their age, sex, and salary is shown in Figure
16.19. Decision trees have proved to be an attractive tool for data mining because
they are simple to implement and their intuitive representation makes the resulting
classification model easy to understand.
Inductive concept learning. It derives a concise, logical description of a
concept from a set of examples.
3. Conceptual clustering. It finds groups or clusters in a set of objects based on
conceptual closeness among objects.
Database-oriented methods use data modeling or database specific heuristics to
exploit the characteristics of data in hand. Some of the commonly used databaseoriented methods used in data mining are:
Attribute-oriented induction. It generalizes primitive, low-level data into
high-level concepts by using conceptual hierarchies.
View Full Document
This document was uploaded on 04/07/2014.
- Spring '14