Computer systems that learn: classification and prediction methods from statistics, neural nets, machine learning, and expert systems
From data mining to knowledge discovery: an overview
Advances in knowledge discovery and data mining
The process of knowledge discovery in databases
Advances in knowledge discovery and data mining
Bayesian classification (AutoClass): theory and results
Advances in knowledge discovery and data mining
Data Mining: An Overview from a Database Perspective
IEEE Transactions on Knowledge and Data Engineering
Applying AI Clustering to Engineering Tasks
IEEE Expert: Intelligent Systems and Their Applications
Scaling the data mining step in knowledge discovery using oceanographic data
IEA/AIE '00 Proceedings of the 13th international conference on Industrial and engineering applications of artificial intelligence and expert systems: Intelligent problem solving: methodologies and approaches
Hi-index | 0.00 |
Knowledge discovery from image data is a multi-stepiterative process. This paper describes the procedure we have usedto develop a knowledge discovery system that classifies regions ofthe ocean floor based on textural features extracted from acousticimagery. The image is subdivided into rectangular cells calledtexture elements (texels); a gray-level co-occurence matrix (GLCM) iscomputed for each texel in four directions. Secondary texturefeatures are then computed from the GLCM resulting in a featurevector representation of each texel instance. Alternatively, aregion-growing approach is used to identify irregularly shapedregions of varying size which have a homogenous texture and for whichthe texture features are computed. The Bayesian classifier Autoclassis used to cluster the instances. Feature extraction is one of themajor tasks in knowledge discovery from images. The initial goal ofthis research was to identify regions of the image characterized bysand waves. Experiments were designed to use expert judgements toselect the most effective set of features, to identify the best texelsize, and to determine the number of meaningful classes in the data.The region-growing approach has proven to be more successful than thetexel-based approach. This method provides a fast and accuratemethod for identifying provinces in the ocean floor of interest togeologists.