Stock trend prediction based on fractal feature selection and support vector machine.

Liping Ni Zhiwei Ni Yazhuo Gao

Published in: Expert Syst. Appl. (2011)

Keyphrases

feature selection
support vector machine
stock market prediction
support vector
multi class
prediction accuracy
small sample
grey model
svm classifier
feature space
classification accuracy
machine learning

selected features
support vector machine svm
text classification
feature set
text categorization
feature extraction
prediction error
stock market
ensemble classifier
parameter optimization
financial time series

classification method
information gain
data sets
kernel function
image compression
feature vectors
radial basis function
prediction model
training data
fractal dimension
feature selection algorithms

training set
mutual information
iterated function systems
automatic relevance determination
stock trading
irrelevant features
knn
ls svm
support vector regression
kernel methods
microarray data