Transductive Inference for Text Classification using Support Vector Machines
Transductive Inference for Text Classification using Support Vector Machines
复制标题
DOI：
--
复制
发表时间：
1999-06
期刊：
影响因子：
--
通讯作者：
T. Joachims
中科院分区：
文献类型：
--
作者：
T. Joachims
本文介绍了用于文本分类的直推式支持向量机（TSVMs）。常规支持向量机（SVM）试图为学习任务引入一般决策函数，而转换支持向量机考虑特定的测试集，并试图最小化仅那些特定示例的误分类。本文分析了为什么支持向量机非常适合文本分类。这些理论结果得到了三个测试集上的实验的支持。实验表明，归纳方法，特别是小的训练集，削减标记的训练样本的数量下降到二十分之一的一些任务有实质性的改善。这项工作还提出了一个算法，用于训练TSVM e（cid：14）-ciently，处理10，000个例子和更多。
This paper introduces Transductive Support Vector Machines (TSVMs) for text classi(cid:12)-cation. While regular Support Vector Machines (SVMs) try to induce a general decision function for a learning task, Transduc-tive Support Vector Machines take into account a particular test set and try to minimize misclassi(cid:12)cations of just those particular examples. The paper presents an analysis of why TSVMs are well suited for text classi(cid:12)cation. These theoretical (cid:12)ndings are supported by experiments on three test collections. The experiments show substantial improvements over inductive methods, espe-ciallyfor smalltraining sets, cutting the number of labeled training examples down to a twentieth on some tasks. This work also proposes an algorithm for training TSVMs e(cid:14)-ciently, handling 10,000 examples and more.