2004 LargeScaleBayesianLogRegForTextCat

• (GenkinLM, 2004) ⇒ A. Genkin, D. Lewis, D. Madigan. (2004). “Large-scale bayesian logistic regression for text categorization.” Technical report, Rutgers University.

Subject Headings:

References

• M. Y. Park and Trevor Hastie. (2006). L1 Regularization Path Algorithm for Generalized Linear Models
  • "Other researchers have implemented algorithms for L1 regularized logistic regression for

diverse applications. For example, Genkin, Lewis & Madigan (2004) proposed an algorithm for L1 regularized logistic regression (for text categorization) in a Bayesian context, in which the parameter of the prior distribution was their regularization parameter. They chose the parameter based on the norm of the feature vectors or through cross-validation, performing a separate optimization for each potential value. Our method of using the solutions for a certain as the starting point for the next, smaller o ers the critical advantage of reducing the number of computations.

---

,

	Author	volume	Date Value	title	type	journal	titleUrl	doi	note	year
2004 LargeScaleBayesianLogRegForTextCat	A. Genkin D. Lewis D. Madigan			Large-scale bayesian logistic regression for text categorization			http://stat.rutgers.edu/~madigan/PAPERS/techno-06-09-18.pdf			2004