Sketches of numeric data series (or time series) are projections of multidimensional vectors onto a space of significantly lower dimensionality with the property to approximately preserve the Euclidean distance at the cost of a minor loss in accuracy. Thus, a similarity query performed on data sketches results in significantly lower response time than the same query performed on the original vectors. This work introduces an approach to estimate the result of a similarity query (that uses sketches) in intuitionistic fuzzy terms, where the degree of uncertainty is mapped to the loss of accuracy traded for performance. The approach is validated through the use of SQL queries of the kind “is x similar to q?” against array-like data sets, whose result set resembles an intuitionistic fuzzy set, where each row is assigned degrees of membership and non-membership.