This paper presents an innovative framework for application of machine learning to development of high-entropy alloys with desired properties. The nature of multi-principal elements, high mixing entropy, and mutual interactions between elements render these alloys with outstanding mechanical and functional properties, such as high hardness and elastic modulus, superior wear resistance, corrosion and temperature resistance, together with appealing electrical and magnetic properties. We present a framework for predicting the composition, yield strength, material ductility, fatigue, fracture toughness, and creep of high-entropy alloys. For each output property of interest, we identify the corresponding driving factors. These input factors may include the material composition, heat treatment, process, microstructure, temperature, strain rate, environment, and testing mode. We then carry out the prediction through a customized averaging process in the space comprising the input parameters. Discussion on incorporation of physics-based models for improved prediction accuracy and optimization of additive manufacturing processes is also included.