A Loss Function Analysis for Classification Methods in Text Categorization

A Loss Function Analysis for Classification Methods in Text Categorization (2003)

Cached

Download Links

[www.cs.cmu.edu]

Other Repositories/Bibliography

DBLP

by Fan Li , Yiming Yang

Citations:

20 - 5 self

Versions

Version History

Metadata Version 2

User correction supplied by SystemCorrections

Datum	Value	Source
TITLE	A Loss Function Analysis for Classification Methods in Text Categorization	user correction - Legacy Corrections
AUTHOR NAME	Fan Li	user correction - Legacy Corrections
AUTHOR NAME	Yiming Yang	user correction - Legacy Corrections
AUTHOR AFFIL	Carnegie Mellon Univ	user correction - Legacy Corrections
AUTHOR ADDR	4502 NSH, 5000 Forbes Avenue, Pittsburgh, PA 15213 USA	user correction - Legacy Corrections
ABSTRACT	This paper presents a formal analysis of popular text classification methods, focusing on their loss functions whose minimization is essential to the optimization of those methods, and whose decomposition into the training-set loss and the model complexity enables cross-method comparisons on a common basis from an optimization point of view. Those methods include Support Vector Machines, Linear Regression, Logistic Regression, Neural Network, Naive Bayes, K-Nearest Neighbor, Rocchio-style and Multi-class Prototype classifiers. Theoretical analysis (including our new derivations) is provided for each method, along with evaluation results for all the methods on the Reuters-21578 benchmark corpus. Using linear regression, neural networks and logistic regression methods as examples, we show that properly tuning the balance between the training-set loss and the complexity penalty would have a significant impact to the performance of a classifier. In linear regression, in particular, the tuning of the complexity penalty yielded a result (measured using macro-averaged F1) that outperformed all text categorization methods ever evaluated on that benchmark corpus, including Support Vector Machines.	user correction - Legacy Corrections
YEAR	2003	user correction - Legacy Corrections
CITATIONS	8 found	ParsCit 1.0