Sediment transport load monitoring is important in civil and environmental engineering fields. Monitoring the total load is difficult, especially because of the cost of the bed load transport measurement. This study proposes estimation models for the suspended load to total load ratio (Fsus) using dimensionless hydro-morphological variables. Two prominent variable combinations were identified using the recursive feature elimination procedure of support vector regression (SVR): (1) W/h, d*, Reh, Frd, and Rew and (2) Reh, Fr, and Frd. The explicit interactions between Fsus and the two combinations were revealed by two modern symbolic regression methods: multi-gene genetic programming and Operon. The five-variable SVR model showed the best performance (R2=0.7722). The target dataset was clustered by applying a self-organizing map and Gaussian mixture model. Through these steps, Reh and Frd are determined as the two most influential variables. Subsequently, the one-at-a-time sensitivity of the input variables of the empirical models was investigated. By referring to the clustering and sensitivity analyses, this study provides physical insights into Fsus controlling relationships. For example, Fsus is proportional to Reh and is inversely related to Frd. The empirical models developed in this study are applicable in practice and easy to implement in other real-time surrogate suspended-sediment monitoring methods, because they only require basic measurable hydro-morphological variables, such as velocity, depth, width, and mean bed material grain size.