We report on extensive experiments using rule-based induction methods for document classification. The goal is to automatically discover patterns in document classifications, potentially surpassing humans who currently read and classify these documents. By using a decision rule model, we induce results in a form compatible with expensive human engineered systems that have recently demonstrated ...

A proposal for text mining as a support for knowledge discovery on biological descriptions is introduced. Our aim is both to sustain the curation of databases and to offer an alternative representation frame for accessing information in the biodiversity domain. We works on raw texts with minimum human intervention, applying natural language processing to integrate linguistic and domain knowledg...

Intelligent geographic information system (IGIS) is one of the promising topics in GIS field. It aims at making GIS tools more sensitive for large volumes of data stored inside GIS systems by integrating GIS with other computer sciences such as Expert system (ES) Data Warehouse (DW), Decision Support System (DSS), or Knowledge Discovery Database (KDD). One of the main branches of IGIS is the Ge...

KDD is a rapidly expanding field with promise for great applicability. Knowledge discovery became the new database technology for the incoming years. The need for automated discovery tools caused an explosion in the number and type of tools available commercially and in the public domain. These requirements encouraged us to propose a new KDD model so called ODBC_KDD(2) described in [39] ."One o...

Traditional pratice in machine learning algorithms involve fixed data sets and static models. Most of the times, all the data is loaded into memory and the learning task is solved by performing multiple scans over the training data. These assumptions fail with the advent of new application areas, like ubiquitous computing, sensor networks, e-commerce, etc., where data flows continuously, eventu...

If it is to qualify as knowledge, a learner's output should be accurate, stable and comprehensible. Learning multiple models can improve signiicantly on the accuracy and stability of single models, but at the cost of losing their comprehensibility (when they possess it, as do, for example, simple decision trees and rule sets). This article proposes and evaluates CMM, a meta-learner that seeks t...

Subgroup discovery can be applied for exploration or descriptive induction in order to discover ”interesting” subgroups of the general population, given a certain property of interest. In domains with available background knowledge, the user usually wants to utilize this to improve the quality of the subgroup discovery results. We describe a knowledge-intensive approach for subgroup discovery u...

Linked Data has been increasing rapidly by publishing machine readable structured data. DBpedia and YAGO are cross-domain data sets, which provide semantic knowledge of things. Although both data sets contain millions of entities, there are still missing knowledge exist in each data set. In this paper, we analyze graph patterns of Linked Data entities to discover missing knowledge in the data s...

A novel knowledge discovery technique using neural networks is presented. A neural network is trained to learn the correlations and relationships that exist in a dataset. The neural network is then pruned and modified to generalize the correlations and relationships. Finally, the neural network is used as a tool to discover all existing hidden trends in four different types of crimes in US citi...

We present a method of prioritizing potential drug targets based on their gene expression “signature”. Primary human pre-cursor neuronal cells were treated with three classes (antidepressant, antipsychotic, opiod receptor agonist) of psychoactive drugs for 24 hours. Microarray technology was used to capture expression of ~11, 000 genes induced by these three categories of drugs. It was demonstr...