Linear and support vector regressions based on geometrical correlation of data

• Main Authors: Kaijun Wang, Junying Zhang, Lixin Guo, Chongyang Tu
• Format: Article
• Language: English
• Published: Ubiquity Press 2007-10-01
• Series: Data Science Journal
• Subjects: Geometrical correlation learning; Geometrical correlation of data; Regression analysis
• Online Access: http://datascience.codata.org/articles/356

Linear regression (LR) and support vector regression (SVR) are widely used in data analysis. Geometrical correlation learning (GcLearn) was proposed recently to improve the predictive ability of LR and SVR through mining and using correlations between data of a variable (inner correlation). This paper theoretically analyzes prediction performance of the GcLearn method and proves that GcLearn LR and SVR will have better prediction performance than traditional LR and SVR for prediction tasks when good inner correlations are obtained and predictions by traditional LR and SVR are far away from their neighbor training data under inner correlation. This gives the applicable condition of GcLearn method.