In this work entitled "A study of some important Applications of Data Mining Techniques - Numerical Investigation and Performance Evaluation" four types of Data Mining Applications are attempted. The first Data Mining Application is to the phenomenon of salt -finger convection in superposed layers. The physical problem is the onset of finger convection in a porous layer underlying a fluid layer in a gravity modulated environment. The Second Data Mining Application is the implementation of three sequential mining algorithms .More emphasis is laid on PLWAP algorithm which is a version of the WAP tree algorithm that assigns unique binary position code to each tree node and performs the header node linkages in the pre-ordered fashion (root, left, right). The application of Genetic Algorithms to unconstrained non linear optimization problems is the Third Data Mining Application. It is observed from the present investigation that the methodology is unique in several respects and very informative. In addition to the above a novel GA (based on clustering) is also discussed.The Fourth Data Mining Application is with regard to Genetic programming and neural networks in medical data mining