Results 221 to 230 of about 313,307 (262)

Feature Construction and Feature Selection in Presence of Attribute Interactions

2009
When used for data reduction, feature selection may successfully identify and discard irrelevant attributes, and yet fail to improve learning accuracy because regularities in the concept are still opaque to the learner. In that case, it is necessary to highlight regularities by constructing new characteristics that abstract the relations among ...
Leila Shila Shafti, Eduardo Pérez
openaire   +1 more source

PSO for feature construction and binary classification

Proceedings of the 15th annual conference on Genetic and evolutionary computation, 2013
In classification, the quality of the data representation significantly influences the performance of a classification algorithm. Feature construction can improve the data representation by constructing new high-level features. Particle swarm optimisation (PSO) is a powerful search technique, but has never been applied to feature construction.
Bing Xue 0001, Mengjie Zhang 0001, Yan Dai 0007, Will N. Browne   +3 more
openaire   +2 more sources

Features in periphrastic constructions

2010
AbstractThis chapter argues that periphrastic constructions can express features which are not part of the content of their elements. This favours an analysis which integrates them in the morphological paradigm. On the other hand, agreement data suggest that a treatment along these lines is not unproblematic.
openaire   +1 more source

A data-driven approach to feature construction

1994
This paper presents a general scheme for feature construction and its application to decision trees. In this scheme, a higher level attribute is constructed from two lower level ones under the guidance of the distributions of the examples from different classes. It can be used with different selective induction algorithms such as decision tree learning,
Jianping Zhang 0003, Hsueh-Hsiang Lu
openaire   +1 more source

Fragmentation problem and automated feature construction

Proceedings Tenth IEEE International Conference on Tools with Artificial Intelligence (Cat. No.98CH36294), 2002
Selective induction algorithms are efficient in learning target concepts but inherit a major limitation each time only one feature is used to partition the data until the data is divided into uniform segments. This limitation results in problems like replication, repetition, and fragmentation.
Rudy Setiono, Huan Liu 0001
openaire   +1 more source

DARA: Data Summarisation with Feature Construction

2008 Second Asia International Conference on Modelling & Simulation (AMS), 2008
This paper addresses the question whether or not the descriptive accuracy of the DARA (Dynamic Aggregation of Relational Attributes) algorithm benefits from the feature construction process. This involves solving the problem of constructing a set of relevant features used to generate patterns representing records in the TF-IDF weighted frequency matrix
openaire   +1 more source

Efficient Training of Evolution-Constructed Features

2015
Evolution-Constructed (ECO) features have been shown to be effective for general object recognition. ECO features use evolution strategies to build series of transforms and thus can be generated automatically without human expert involvement. We improved on our successful ECO features algorithm by reducing their dimensions before putting them into the ...
Meng Zhang, Dah-Jye Lee
openaire   +1 more source

Feature Construction in Structural Decision Trees

1991
STRUCT is a system that learns structural decision trees from positive and negative examples. The algorithm uses a modification of Pagallo and Haussler's FRINGE algorithm to construct new features in a first-order representation. Experiments compare the different feature construction strategies.
Larry Watanabe, Larry A. Rendell
openaire   +1 more source

Constructing invariant features by averaging techniques

Proceedings of the 12th IAPR International Conference on Pattern Recognition (Cat. No.94CH3440-5), 2002
The paper presents algorithms for the construction of features which are invariant with respect to a given transformation group. The methods are based on integral calculus and are applicable to parametric groups (Lie groups) and finite groups as well. To illustrate the concepts the author discusses in detail how to construct invariant features for the ...
openaire   +1 more source

Learning Polynomial Functions by Feature Construction

1991
We present a method for learning higher-order polynomial functions from examples using linear regression and feature construction. Regression is used on a set of training instances to produce a weight vector for a linear function over the feature set.
Richard S. Sutton, Christopher J. Matheus   +1 more
openaire   +1 more source