Frequently we become amazed with the increasing number of problems to be solved that fiourish while facing daily activities. Often, related to these problems we llave also an incredible amount oí data. Since we cannot allways afford time and resources to sol ve them, we keep on gathering and storing data in large databases, widening the gap between raw and interpreted data. At this point we should refiect about Polya's maxima "A great discovery solves a great problem" and realize that databases encompass the knowledge necessary for guiding the decision making process. The question that remains is how to organize and explore this knowledge. This paper presents sorne approaches to knowledge discovery in databases íound in the literature, analyzing issues in classifying and clustering large data sets.