The effective application of a Decision Tree (DT) process is beset with many difficult and technical decisions about the choice of algorithms, parameters, evaluation, etc. Therefore, we propose assistance by using ontologies for addressing the above-mentioned challenges that face the non-specialist DT miner (person). Ontologies have been used in various research areas such as computer science, including data mining tools. In this paper, we propose the realisation of a domain ontology for DT OntoDTA to empower the non-specialist DT miner throughout the key phases of the DT process. OntoDTA ontology contains the knowledge of DT and provides a common terminology that can be shared and processed by DT miners.

Abstract-The domain med ical and public health remains the principal preoccupation of all world population. It makes recourse at several means fro m various disciplines, including for instance epidemiology, to help clinicians in decision processes. This paper p roposes an Assistance Platform fo r Epidemiological Searches and Surveillance (APESS) for service-oriented data min ing in the field of epidemiology. The main aim of the present platform is to build a system that enables extract ing predictive ru les, flexib le and scalable for aid in decision-making by trades' experts. Results showed that the current system provides prediction models of chronic d iseases (epidemio logical prediction ru les), using classification algorithms.

No abstract

For a deeper and richer analytic processing of medical datasets, feature selection aims to eliminate redundant and irrelevant features from the data. While filter has been touted as one of the simplest methods for feature selection, its applications have generally failed to identify and deal with embedded similarities among features. In this research, a hybrid approach for feature selection based on combining the filter method with the hierarchical agglomerative clustering method is proposed to eliminate irrelevant and redundant features in four medical datasets. A formal evaluation of the proposed approach unveils major improvements in the classification accuracy when results are compared to those obtained via only the applications of the filter methods and/or more classical-based feature selection approaches.