Data Mining We have developed a means of combining data from different sources for co-existence in a single database for data mining.  The process involves standardizing the input at each of the sources.  XML is used for flexibility, and it easily allows each site to mark the fields which are important and also to define their corresponding fields in the underlying database.  This process is called source labeling since only the information taken directly from the source is marked.  Care is taken to ensure that the physical field names match from site to site to avoid redundancy.   Once the input from the data sources is standardized, a pre-processing step is taken at data mining.  We refer to this step as derivative labeling, as fields are added which are based on the information contained in the source fields.  next Please contact us for more information.

		home	search			designed by wbdigital