The 25th International Conference on Machine Learning (ICML 2008)

Local Likelihood Modeling of Temporal Text Streams

author: Guy Lebanon, Purdue University

Description

Temporal text data is often generated by a time-changing process or distribution. Such a drift in the underlying distribution cannot be captured by stationary likelihood techniques. We consider the application of local likelihood methods to generative and conditional modeling of temporal document sequences. We examine the asymptotic bias and variance and present an experimental study using the RCV1 dataset containing a temporal sequence of Reuters news stories.

You might be experiencing some problems with Your Video player.

*Slides*
0:00	Local Likelihood Modeling of Temporal Text Streams
0:02	Concept Drift - 1
1:14	Concept Drift - 2
1:33	Concept Drift - 3
1:47	Concept Drift - 4
2:41	Concept Drift - 5
3:48	Concept Drift in pt (x\|y )
4:58	Local Likelihood - 1
5:25	Local Likelihood - 2
5:44	Local Likelihood - 3
6:02	Local Likelihood - 4
6:16	Local Likelihood - 5
6:21	Local Likelihood - 6
6:36	Sampling Density
8:22	Global and Local Models - 1
8:49	Global and Local Models - 2
9:03	Global and Local Models - 3
9:23	Global and Local Models - 4
9:38	Global and Local Models - 5
9:45	Global and Local Models - 6
10:01	Global and Local Models - 7
10:29	Local Likelihood - 7
10:34	Local Likelihood - 8
11:05	Local Likelihood - 9
11:43	Local Likelihood - 10
12:09	Smoothing Kernel - 1
12:46	Smoothing Kernel - 2
12:54	Smoothing Kernel - 3
13:34	Smoothing Kernel - 4
14:18	Local Likelihood for n-Grams - 1
14:32	Local Likelihood for n-Grams - 2
14:51	Precise Bias Variance Analysis
15:26	Asymptotic Bias Variance Analysis - 1
15:58	Asymptotic Bias Variance Analysis - 2
16:17	Asymptotic Bias Variance Analysis - 3
16:37	Asymptotic Bias Variance Analysis - 4
17:14	Asymptotic Bias Variance Analysis - 1
17:16	Asymptotic Bias Variance Analysis - 5
18:25	Bandwidth Selection - 1
18:41	Bandwidth Selection - 2
18:57	Bandwidth Selection - 3
19:23	Bandwidth Selection - 4
19:49	Bandwidth Selection - 5
21:00	Local Likelihood for Logistic Regression - 1
21:08	Local Likelihood for Logistic Regression - 2
21:23	Local Likelihood for Logistic Regression - 3
21:35	Local Likelihood for Logistic Regression - 4
21:46	Local Likelihood for Logistic Regression - 5
23:12	Discussion - 1
23:30	Discussion - 2
23:38	Discussion - 3
24:23	Discussion - 4
24:56	Concept Drift in pt (x\|y )
25:32	- Questions

Lecture rating

People found this lecture:
Worth seeing
because it is:
Valuable and informative
Well presented
Easily understandable
Acceptably recorded
You need to login to cast your vote.

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.