An important component of Gravitional wave data analysis is the search for unmodelled quasiperiodic bursts of gravitional waves. An important family of algorithms for this are time frequency methods wherein one constructs a time frequency map of the data and then pick out "features" which could be signatures of a gravitational wave burst. One such algorithm is the Hierarchical Algorithm for Curves and Ridges(HACR). Such algorithms however have a large number of parameters which are required to be optimized for specific searches. These include the parameters used to construct the time frequency map and the the parameters which distinguish signals on the map from the background noise. In this talk I present a methodology to identify optimal parameters.