Australasian Conference on Artificial Intelligence 2010

16 papers
The 2010 Australasian Conference on Artificial Intelligence featured 18 papers focusing on machine learning, evolutionary computation, and data processing. A central theme involved genetic programming, specifically addressing classification challenges with unbalanced data. Other evolutionary methods were represented through research on particle swarm optimisation for edge and corner extraction and the development of encoding schemes for cooperative coevolutionary feedforward neural networks. Algorithmic research during this edition included the introduction of Newton Trees and the Incremental Projection Vector Machine, a one-stage learning algorithm intended for high-dimension, large-sample datasets. Dimensionality reduction was explored through visualization techniques such as L1SNE, while other studies reviewed modern logistic regression methods for small and medium sample sizes. Prolific authors contributing to these themes included Mengjie Zhang, Mark Johnston, and Fernando Martínez-Plumed. Collectively, the research emphasized the refinement of classification strategies and the application of optimization techniques to complex data structures.

Top Authors

AuthorPapers
Mengjie Zhang5
Fernando Martínez-Plumed1
Jose Hernandez-Orallo1
Junbin Gao1
Jun Liu1
Kai Ming Ting1
Kazunori Matsumoto1
Maurice Pagnucco1
Qinghua Zheng1

Papers

Title
A Comparative Study of a Practical Stochastic Clustering Method with Traditional Methods
A Comparison of Classification Strategies in Genetic Programming with Unbalanced Data
A Gradient-Based Metric Learning Algorithm for k-NN Classifiers
An Encoding Scheme for Cooperative Coevolutionary Feedforward Neural Networks
Database Normalization as a By-product of Minimum Message Length Inference
Dimensionality Reduction for Classification through Visualisation Using L1SNE
Edge and Corner Extraction Using Particle Swarm Optimisation
Evaluating Sequential Single-Item Auctions for Dynamic Task Allocation
Hazardous Document Detection Based on Dependency Relations and Thesaurus
Hybrid Particle Swarm Optimisation Algorithms Based on Differential Evolution and Local Search
Incremental Projection Vector Machine: A One-Stage Learning Algorithm for High-Dimension Large-Sample Dataset
Newton Trees
Review of Modern Logistic Regression Methods with Application to Small and Medium Sample Size Problems
Sampling Methods in Genetic Programming for Classification with Unbalanced Data
The Behaviour of the Akaike Information Criterion When Applied to Non-nested Sequences of Models
Two Methods for Constructing Horn Contractions