Feature Shaping for Linear SVM Classifiers

Forman, George ; Scholz, Martin ; Rajaram, Shyamsundar

HP Development Company

关键词: text classification machine learning;    feature weighting;    feature scaling;    SVM;   

RP-ID  :  HPL-2009-31R1

学科分类：计算机科学（综合）

美国|英语

来源: HP Labs

【 摘 要 】

Linear classifiers have been shown to be effective for many discrimination tasks. Irrespective of the learning algorithm itself, the final classifier has a weight to multiply by each feature. This suggests that ideally each input feature should be linearly correlated with the target variable (or anti- correlated), whereas raw features may be highly non- linear. In this paper, we attempt to re-shape each input feature so that it is appropriate to use with a linear weight and to scale the different features in proportion to their predictive value. We demonstrate that this pre-processing is beneficial for linear SVM classifiers on a large benchmark of text classification tasks as well as UCI datasets.