Abstract | ||
---|---|---|
Feature selection is a critical preprocessing step in machine learning. It contributes to cost-effective model building and improvement of model prediction performance. Generally, a feature selection algorithm requires a dependency measure and a search strategy. Extant dependency measures are mostly based on pair-wise correlation analysis, which cannot detect feature interaction. To overcome this problem, we developed a unified dependency criterion called inference correlation. The inference correlation between a set of predictor variables and a response variable can be efficiently calculated. The variables could be discrete, continuous, or mixed. Therefore, inference correlation can be applied to select features for both classification and regression problems. A feature selection algorithm using sequential floating forward search based on inference correlation is presented. Experiments of the algorithm on synthetic datasets and real-world problems confirm the effectiveness of the feature selection approach when compared to extant feature selection methods. |
Year | DOI | Venue |
---|---|---|
2011 | 10.3233/IDA-2010-0473 | Intell. Data Anal. |
Keywords | Field | DocType |
feature selection,feature interaction,unified dependency criterion,feature selection algorithm,extant dependency measure,feature selection method,inference correlation,dependency measure,feature selection approach,pair-wise correlation analysis | k-nearest neighbors algorithm,Dimensionality reduction,Feature selection,Pattern recognition,Regression,Inference,Computer science,Feature (computer vision),Model building,Preprocessor,Artificial intelligence,Machine learning | Journal |
Volume | Issue | ISSN |
15 | 3 | 1088-467X |
Citations | PageRank | References |
8 | 0.55 | 16 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Dengyao Mo | 1 | 15 | 1.75 |
Samuel H. Huang | 2 | 193 | 19.64 |