Transcript Machine Learning - University of Houston

Ricardo Vilalta
Dept. of Computer Science
University of Houston
September, 2013
Objectives
To extend traditional data analysis techniques with
specialized analysis tools.
To develop new data mining techniques with additional
levels of efficiency.
I have the privilege of working with the following students:
Bachelor’s in Computer Science
• Dutton, Benjamin
Master's in Computer Science
• Ahmed, Owais
• Boumber, Dainis
• Sui, Bangsheng
Doctorate in Computer Science
• Dhar Gupta, Kinjal
• Toti, Giulia
• Valerio, Roberto
Field of Study
Artificial Intelligence
Planning
Knowledge
Representation
Search
Machine Learning
Clustering
Robotics
Reinforcement
Learning
Genetic Algorithms
Classification
Multidisciplinary Field
Probability &
Statistics
Computational
Complexity
Theory
Artificial
Intelligence
Neurobiology
Machine
Learning
Information
Theory
Philosophy
Meta-learning
Model
Selection
Physics and
Astronomy
Feature Selection
Classification
Regression
Clustering
The Connection to Data Mining
Identifying Sources of Ionizing Radiation
Identifying Sources of Ionizing Radiation
Classifying Cepheid Stars
Mean Magnitude Vs Log Period In I band
16
17
Cepheids Fundamentals of LMC
Cepheids First Overtones of LMC
Cepheids of M33
18
19
Mag
20
21
22
23
24
25
-1
-0.5
0
0.5
1
LogPeriod
1.5
2
2.5
Prediction of stuck pipes during drilling using
machine learning techniques
Time-Based Data
Prediction
The goal is to predict confidently within a time window the
Safe15time
window for prediction
occurrence of a stuck pipe. Warning Window:
minutes
Stuck Pipe Event
THANK YOU