22nd Annual International Conference on Machine Learning (ICML), Bonn 2005

BayesANIL - A Bayesian Model for Handling Approximate, Noisy or Incomplete Labeling in Text Classification

author: Ganes Ramakrishnan, IBM
published: Feb. 25, 2007, recorded: August 2005, views: 108

Slides

Slides
0:06	BayesANIL A Bayesian Model for Handling Approximate, Noisy or Incomplete Labeling in Text Classification
0:25	Outline
0:37	Motivation - hurdles in supervised learning of text classifiers
1:29	Related work
2:52	Related work (contd.)
4:00	What we propose
5:09	Role of BayesANIL in text classification
7:27	The BayesANIL model : notations
8:30	The BayesANIL model : notations…
9:47	The BayesANIL model: Objective function
10:33	The BayesANIL model: E and M Steps
11:09	The Algorithm
12:42	Re-estimating the empirical distribution
14:09	Utilizing parameters of BayesANIL in NB
14:58	Utilizing parameters of BayesANIL in SVM
16:00	Experiments and Results
17:03	Experiments and Results: Supervised
18:23	Experiments and Results: Labeled-unlabeled
19:20	Experiments and Results: Access to unlabeled for WebKB
20:18	Experiments and Results: Access to unlabeled for 20 Newsgroups
21:42	Experiments and Results: Noisy Labels
23:21	Comparison with results as reported by (Bing Liu et al 2003)
24:05	Experiments and Results: Notion of Support
25:13	Summary
26:25	Future work

Report a problem or upload files

If you have found a problem with this lecture or would like to send us extra material, articles, exercises, etc., please use our ticket system to describe your request and upload the data.
Enter your e-mail into the 'Cc' field, and we will keep you updated with your request's status.

Lecture popularity: You need to login to cast your vote.