Pesticides in Drinking Water and Public Health: Modeling Using Artificial Intelligence Techniques

Abstract

Abstract. The causes of chronic diseases, including cancers, are extremely complex. Throughout their lives, individuals who differ in genetic makeup and sensitivity are exposed to a wide variety of carcinogens. Some chemicals by themselves are safe but can act as synergists or promoters together with other chemicals to cause the disease. The problem of the harmful effects of pesticides is the long-term consequences. If the acute effects are well known, the long-term projections remain to be elucidated. Surface water has a direct influence on the degree of contamination by pesticides in groundwater. These are contaminated by significant infiltrations. In order to establish a direct relationship between the possible chronic effects of pesticides on public health, it is essential to consider a set of factors that intervene in the process. Some factors are related to climatic conditions, others to the nature of the soil and others to the nature of the pesticide used. Other factors are related to the people who consume these waters in terms of period, age, sex ... etc. Making a classical mathematical analysis becomes very difficult and even impossible. We propose in this study the analysis of these variables by the techniques of artificial intelligence mainly fuzzy inference. The variables analyzed are considered as fuzzy variables. This makes it possible to take care of these uncertainties and vagueness. A database is established, the resulting algorithm will allow us to read instantly the outcome which expresses the degree of contamination on public health from the introduction of the random variables to the input of the system.