The physicochemical properties of calcined clay have a dominant effect on the rheological properties, such as yield stress and plastic viscosity, of limestone calcined clay cements (LC3). However, understanding is lacking on how other factors—including limestone particle size and gypsum addition rate—influence the rheology of LC3, particularly with different calcined clay contents. Here, water-to-solid ratio (W/S), constituent mass ratios (PC:Metakaolin:Limestone), limestone particle size and gypsum content were varied and rheological properties were assessed both with a rheometer and mini slump test in mixes prepared with consistent shearing history. From these data, a machine learning model for LC3 yield stress and plastic viscosity was developed using a support vector machine algorithm and the correlations were established between the spread diameter from the mini slump test and the yield stress and plastic viscosity measured by the rheometer. At constant solid volume fraction, higher yield stress and plastic viscosity result from a greater proportion of metakaolin and finer limestone particle sizes, but higher gypsum addition rates can decrease both the yield stress and plastic viscosity. The results show that a single compositional parameter (e.g., solid volume fraction, metakaolin proportion) is not adequate to predict LC3 rheology. Therefore, given these complexities, it is suggested that a data analytics approach offers an advantage because it combines numerous compositional parameters to make accurate predictions.