Computational Intelligence in Survival Analysis

Computational Intelligence in Survival Analysis

Malgorzata Kretowska (Bialystok University of Technology, Poland)
Copyright: © 2014 |Pages: 11
DOI: 10.4018/978-1-4666-5202-6.ch044
OnDemand PDF Download:
$30.00
List Price: $37.50

Chapter Preview

Top

Background

In this section short introduction to survival data, as well as to artificial neural networks and tree based models is provided.

Key Terms in this Chapter

Computational Intelligence: A term which describes methods, which are able to model complex relationships in the data.

Artificial Neural Networks: Mathematical models inspired by biological neural networks. They consist of several layers of artificial neurons. The structure and weights of connection between neurons are changing during learning phase. Neural networks are able to model complex relationships between input and output data.

Bagging Predictor: A set of single models (e.g. trees) inducted on the base of bootstrap samples drawing with replacement from the original data.

Censored Observation: The object from survival data which do not contain exact information of failure occurrence. For such observation we only know that the failure time is not less than its follow-up time.

Survival Trees: Tree-based models which are developed to analyze survival data. They usually cope with censored observations and, besides some common statistics as median survival time, they often return survival or hazard functions.

Artificial Neuron: Mathematical model of the biological neuron. It calculates the weighed sum of input signals ( x ) and returns the value of activation function.

Complete Chapter List

Search this Book:
Reset