Aluminum alloys are widely used for structural applications due to their excellent strength-to-weight ratio and corrosion resistance. In this study, a dataset on composition, heat-treatment processes and mechanical properties of 1572 aluminum alloys are collected, and various models with an array of algorithms are prepared for their fatigue strength prediction. The best performing model for fatigue strength prediction using composition and heat-treatment data has R2 scores of 0.91. The accuracy improved to R2 scores of 0.98 after including tensile strength, ductility, and hardness, as features in the training dataset. The effects of mechanical properties on the prediction model are also studied with the help of correlation heatmaps and sequential feature selectors. The results from this analysis are compared with the effect of heat-treatment (ageing, hardening, tempering, annealing) and mechanical properties on fatigue strength from conventional viewpoints. Although these data-driven prediction models cannot pinpoint the characteristics of the superior alloys, they can help in narrowing down the large search space.