The 7th International Symposium on Intelligent Data Analysis

Tree Augmented Naive Bayes for Regression Using Mixtures of Truncated Exponentials: Application to Higher Education Management

coauthor: Antonio Salmerón, University of Almería

Description

In this paper we explore the use of Tree Augmented Naive Bayes (TAN) in regression problems where some of the independent variables are continuous and some others are discrete. The proposed solution is based on the approximation of the joint distribution by a Mixture of Truncated Exponentials (MTE). The construction of the TAN structure requires the use of the conditional mutual information, which cannot be analytically obtained for MTEs. In order to solve this problem, we introduce an unbiased estimator of the conditional mutual information, based on Monte Carlo estimation. We test the performance of the proposed model in a real life context, related to higher education management, where regression problems with discrete and continuous variables are common. This work has been supported by the Spanish Ministry of Education and Science, project TIN2004-06204-C03-01 and by Junta de Andalucía, project P05-TIC-00276.

Categories

Top: Computer Science: Machine Learning: Bayesian Learning
Top: Computer Science: Machine Learning: Regression

You might be experiencing some problems with Your Video player.

*Slides*
0:00	Tree Augmented Naive Bayes for Regression Using Mixtures of Truncated Exponentials: Application to Higher Education Management
1:52	Outline pt 1
2:06	Outline pt 2
2:14	Outline pt 3
2:26	Outline pt 4
2:44	Outline pt 5
2:57	Outline pt 6
3:00	Motivation pt 1
3:21	Motivation pt 2
4:17	Motivation pt 3
5:09	Motivation pt 4
5:30	Bayesian Networks pt 1
6:17	Bayesian Networks pt 2
6:46	Bayesian Networks for Classification pt 1
6:57	Bayesian Networks for Classification pt 2
7:43	Bayesian Networks for Classification pt 3
7:53	Bayesian Networks for Classification pt 4
8:03	Bayesian Networks for Classification pt 5
8:09	Bayesian Networks for Classification pt 6
8:32	Bayesian Networks for Classification pt 7
8:55	Bayesian Networks for Regression pt 1
9:50	Bayesian Networks for Regression pt 2
10:32	The Model Proposed by Frank et al. pt 1
10:40	The Model Proposed by Frank et al. pt 2
10:44	The Model Proposed by Frank et al. pt 3
10:53	The Model Proposed by Frank et al. pt 4
10:54	The Model Proposed by Frank et al. pt 5
11:29	The Model Proposed by Morales et al. pt 1
11:30	The Model Proposed by Morales et al. pt 2
11:39	The Model Proposed by Morales et al. pt 3
11:43	The Model Proposed by Morales et al. pt 4
11:50	The MTE Model pt 1
12:38	The MTE Model pt 2
12:52	The MTE Model pt 3
13:48	Increasing the Structural Complexity: TAN Models
14:33	Constructing a TAN Regression Model pt 1
14:50	Constructing a TAN Regression Model pt 2
15:08	Constructing a TAN Regression Model pt 3
15:19	Constructing a TAN Regression Model pt 4
15:29	Constructing a TAN Regression Model pt 5
15:31	Constructing a TAN Regression Model pt 6
15:43	Computing the Conditional Mutual Information pt 1
15:58	Computing the Conditional Mutual Information pt 2
16:04	Computing the Conditional Mutual Information pt 3
16:38	Case Studies pt 1
17:26	Case Studies pt 2
17:36	Case Studies pt 3
17:52	Case Studies pt 4
17:54	Case Studies pt 5
18:21	Results pt 1
18:22	Results pt 2
19:00	Conclusions pt 1
19:01	Conclusions pt 2
19:03	Conclusions pt 3
19:04	Conclusions pt 4
19:05	Conclusions pt 5
19:06	Conclusions pt 6

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.