Machine Learning
A complete fuzzy decision tree technique
Fuzzy Sets and Systems - Theme: Learning and modeling
FuzzyShell: a large-scale expert system shell using fuzzy logic for uncertainty reasoning
IEEE Transactions on Fuzzy Systems
| Hi-index | 0.00 |
The poor characterisation of contaminated soils is likely to result in high costs, restricted choice in landfill disposal sites and future environmental impact. This makes that big quantities of soils are still waiting to be recovered. Until now the tools used to detect contaminated soil have been very generic, without any criteria of prioritization. Usually, simulation studies are used to classify contaminated soils, however these systems need a large quantity of data that is difficult to obtain and manage, which means that the results obtained are subject to large uncertainty. Recently, Artificial Intelligence techniques have been used to tackle this problem. In this work we propose the use of Fuzzy Expert Systems to classify the soils. Classical decision rules have shown to be interpretable, efficient, problem independent and able to treat large scale applications, but they are also recognised as highly unstable classifiers with respect to minor perturbations in the training data. In our problem, the data is subject to uncertainty, for this reason we propose the fuzzyfication of the variables. In the study many different variables have been taken into account: Physic and Chemical characteristics of the soil and pollutants, toxicological properties, and environment and social conditions. After applying Fuzzy Expert Systems at different levels, we obtain a ranking of the soils according to their risk of contamination. The results have been contrasted with another Multicriteria Decision Making system.