The time dimension is a unique and powerful dimension in every enterprise data. In dynamic application such as financial and medical applications representing data as it changes overtime is a common problem. There are diverse applications that require tracing the changes of contents of a data element as time passes. The ability to reason about time and temporal relation is fundamental to almost any intelligent entity that needs to make a decision. Temporal reasoning is a tool to enhance other types of reasoning. Many reasoning tasks such as planning, understanding or diagnosis have an aspect of time. Time-oriented data mining, or knowledge discovery in time-oriented databases, refers to the extraction of implicit knowledge, temporal relations, or other patterns not explicitly stored in time-oriented databases. This research investigates and contributes to the accommodation of temporal semantics within the domain of data mining. It uses the outcome to discover knowledge from medical data where the history of data is very important and discovery of patterns of data over time is crucial.