Ebook: Learning from Data: Artificial Intelligence and Statistics V

Ten years ago Bill Gale of AT&T Bell Laboratories was primary organizer of the first Workshop on Artificial Intelligence and Statistics. In the early days of the Workshop series it seemed clear that researchers in AI and statistics had common interests, though with different emphases, goals, and vocabularies. In learning and model selection, for example, a historical goal of AI to build autonomous agents probably contributed to a focus on parameter-free learning systems, which relied little on an external analyst's assumptions about the data. This seemed at odds with statistical strategy, which stemmed from a view that model selection methods were tools to augment, not replace, the abilities of a human analyst. Thus, statisticians have traditionally spent considerably more time exploiting prior information of the environment to model data and exploratory data analysis methods tailored to their assumptions. In statistics, special emphasis is placed on model checking, making extensive use of residual analysis, because all models are 'wrong', but some are better than others. It is increasingly recognized that AI researchers and/or AI programs can exploit the same kind of statistical strategies to good effect. Often AI researchers and statisticians emphasized different aspects of what in retrospect we might now regard as the same overriding tasks.

---

This volume contains a revised collection of papers originally presented at the Fifth International Workshop on Artificial Intelligence and Statistics in 1995. The topics represented in this collection of 42 papers are diverse and include natural language applications, causality and graphical models, classification, learning, knowledge discovery, and exploratory data analysis. The papers illustrate the rich possibilities for interdisciplinary study at the interface of artificial intelligence and statistics.

---

This volume contains a revised collection of papers originally presented at the Fifth International Workshop on Artificial Intelligence and Statistics in 1995. The topics represented in this collection of 42 papers are diverse and include natural language applications, causality and graphical models, classification, learning, knowledge discovery, and exploratory data analysis. The papers illustrate the rich possibilities for interdisciplinary study at the interface of artificial intelligence and statistics.
Content:
Front Matter....Pages i-xii
Front Matter....Pages 1-1
Two Algorithms for Inducing Structural Equation Models from Data....Pages 3-12
Using Causal Knowledge to Learn More Useful Decision Rules From Data....Pages 13-22
A Causal Calculus for Statistical Research....Pages 23-33
Likelihood-based Causal Inference....Pages 35-44
Front Matter....Pages 45-45
Ploxoma: Testbed for Uncertain Inference....Pages 47-57
Solving Influence Diagrams Using Gibbs Sampling....Pages 59-68
Modeling and Monitoring Dynamic Systems by Chain Graphs....Pages 69-77
Propagation of Gaussian belief functions....Pages 79-88
On Test Selection Strategies for Belief Networks....Pages 89-98
Representing and Solving Asymmetric Decision Problems Using Valuation Networks....Pages 99-108
A Hill-Climbing Approach for Optimizing Classification Trees....Pages 109-117
Front Matter....Pages 119-119
Learning Bayesian Networks is NP-Complete....Pages 121-130
Heuristic Search for Model Structure: the Benefits of Restraining Greed....Pages 131-142
Learning Possibilistic Networks from Data....Pages 143-153
Detecting Imperfect Patterns in Event Streams Using Local Search....Pages 155-164
Structure Learning of Bayesian Networks by Hybrid Genetic Algorithms....Pages 165-174
An Axiomatization of Loglinear Models with an Application to the Model-Search Problem....Pages 175-184
Detecting Complex Dependencies in Categorical Data....Pages 185-195
Front Matter....Pages 197-197
A Comparative Evaluation of Sequential Feature Selection Algorithms....Pages 199-206
Classification Using Bayes Averaging of Multiple, Relational Rule-based Models....Pages 207-217
Front Matter....Pages 197-197
Picking the Best Expert from a Sequence....Pages 219-227
Hierarchical Clustering of Composite Objects with a Variable Number of Components....Pages 229-238
Searching for Dependencies in Bayesian Classifiers....Pages 239-248
Front Matter....Pages 249-249
Statistical Analysis of Complex Systems in Biomedicine....Pages 251-258
Learning in Hybrid Noise Environments Using Statistical Queries....Pages 259-270
On the Statistical Comparison of Inductive Learning Methods....Pages 271-279
Dynamical Selection of Learning Algorithms....Pages 281-290
Learning Bayesian Networks Using Feature Selection....Pages 291-300
Data Representations in Learning....Pages 301-310
Front Matter....Pages 311-311
Rule Induction as Exploratory Data Analysis....Pages 313-322
Non-Linear Dimensionality Reduction: A Comparative Performance Analysis....Pages 323-331
Omega-Stat: An Environment for Implementing Intelligent Modeling Strategies....Pages 333-342
Framework for a Generic Knowledge Discovery Toolkit....Pages 343-352
Control Representation in an EDA Assistant....Pages 353-362
Front Matter....Pages 363-363
A Further Comparison of Simplification Methods for Decision-Tree Induction....Pages 365-374
Robust Linear Discriminant Trees....Pages 375-385
Tree Structured Interpretable Regression....Pages 387-398
An Exact Probability Metric for Decision Tree Splitting....Pages 399-410
Front Matter....Pages 411-411
Two Applications of Statistical Modelling to Natural Language Processing....Pages 413-421
A Model for Part-of-Speech Prediction....Pages 423-432
Front Matter....Pages 411-411
Viewpoint-Based Measurement of Semantic Similarity between Words....Pages 433-442
Part-of-Speech Tagging from “Small” Data Sets....Pages 443-450
Back Matter....Pages 451-452

---

This volume contains a revised collection of papers originally presented at the Fifth International Workshop on Artificial Intelligence and Statistics in 1995. The topics represented in this collection of 42 papers are diverse and include natural language applications, causality and graphical models, classification, learning, knowledge discovery, and exploratory data analysis. The papers illustrate the rich possibilities for interdisciplinary study at the interface of artificial intelligence and statistics.
Content:
Front Matter....Pages i-xii
Front Matter....Pages 1-1
Two Algorithms for Inducing Structural Equation Models from Data....Pages 3-12
Using Causal Knowledge to Learn More Useful Decision Rules From Data....Pages 13-22
A Causal Calculus for Statistical Research....Pages 23-33
Likelihood-based Causal Inference....Pages 35-44
Front Matter....Pages 45-45
Ploxoma: Testbed for Uncertain Inference....Pages 47-57
Solving Influence Diagrams Using Gibbs Sampling....Pages 59-68
Modeling and Monitoring Dynamic Systems by Chain Graphs....Pages 69-77
Propagation of Gaussian belief functions....Pages 79-88
On Test Selection Strategies for Belief Networks....Pages 89-98
Representing and Solving Asymmetric Decision Problems Using Valuation Networks....Pages 99-108
A Hill-Climbing Approach for Optimizing Classification Trees....Pages 109-117
Front Matter....Pages 119-119
Learning Bayesian Networks is NP-Complete....Pages 121-130
Heuristic Search for Model Structure: the Benefits of Restraining Greed....Pages 131-142
Learning Possibilistic Networks from Data....Pages 143-153
Detecting Imperfect Patterns in Event Streams Using Local Search....Pages 155-164
Structure Learning of Bayesian Networks by Hybrid Genetic Algorithms....Pages 165-174
An Axiomatization of Loglinear Models with an Application to the Model-Search Problem....Pages 175-184
Detecting Complex Dependencies in Categorical Data....Pages 185-195
Front Matter....Pages 197-197
A Comparative Evaluation of Sequential Feature Selection Algorithms....Pages 199-206
Classification Using Bayes Averaging of Multiple, Relational Rule-based Models....Pages 207-217
Front Matter....Pages 197-197
Picking the Best Expert from a Sequence....Pages 219-227
Hierarchical Clustering of Composite Objects with a Variable Number of Components....Pages 229-238
Searching for Dependencies in Bayesian Classifiers....Pages 239-248
Front Matter....Pages 249-249
Statistical Analysis of Complex Systems in Biomedicine....Pages 251-258
Learning in Hybrid Noise Environments Using Statistical Queries....Pages 259-270
On the Statistical Comparison of Inductive Learning Methods....Pages 271-279
Dynamical Selection of Learning Algorithms....Pages 281-290
Learning Bayesian Networks Using Feature Selection....Pages 291-300
Data Representations in Learning....Pages 301-310
Front Matter....Pages 311-311
Rule Induction as Exploratory Data Analysis....Pages 313-322
Non-Linear Dimensionality Reduction: A Comparative Performance Analysis....Pages 323-331
Omega-Stat: An Environment for Implementing Intelligent Modeling Strategies....Pages 333-342
Framework for a Generic Knowledge Discovery Toolkit....Pages 343-352
Control Representation in an EDA Assistant....Pages 353-362
Front Matter....Pages 363-363
A Further Comparison of Simplification Methods for Decision-Tree Induction....Pages 365-374
Robust Linear Discriminant Trees....Pages 375-385
Tree Structured Interpretable Regression....Pages 387-398
An Exact Probability Metric for Decision Tree Splitting....Pages 399-410
Front Matter....Pages 411-411
Two Applications of Statistical Modelling to Natural Language Processing....Pages 413-421
A Model for Part-of-Speech Prediction....Pages 423-432
Front Matter....Pages 411-411
Viewpoint-Based Measurement of Semantic Similarity between Words....Pages 433-442
Part-of-Speech Tagging from “Small” Data Sets....Pages 443-450
Back Matter....Pages 451-452
....