Abstract:

Classification process is characterized by identifying clusters within multidimensional data based on a similarity measure. Hence different researchers from several domains are working on the classification problem. The situation is more and more complex and needs powerful techniques for information retrieval especially on a domain that attracts a lot of attention from the data mining community: the biomedical domain. So, this paper attempts to address the problem of classification of multilingual medical documents. In this overview, several types of research using different classification approaches have been presented. A comparison between these approaches highlighted their negative and positive points.