Visualisation as a tool for data analysis

Obviously, the simple method does not decrease the amount of data channelled to the screen. It is quite possible that the numerous less important nodes make it impossible to observe intriguing results in the centre of the area. The objective of more sophisticated techniques is to provide the user with tools to answer his specific questions. In this way, visualisation is a powerful method of data analysis. The most straightforward idea is to display the mesh and thus the associated data only partially. At the same time, it is inconvenient to be restricted by the discrete nature of the elements, and questions the user asks usually concern the continuous result, not values at specified points. Theoretically, we should use the shape functions of the finite element approximation to obtain the values, but it is sufficient to use linear interpolation. In fact, the two are exactly the same in the case of the generally used first-order weighting functions.