The newly developed analysis method for nanoindentation load-displacement curves are focused on not only obtaining elastic modulus and hardness values but also other mechanical properties, such as yield strength and strain hardening properties. Dao et al. developed a forward and reverse algorithm to extract the elasto-plastic properties of materials from the load-displacement curves obtained in nanoindentation test. These algorithms were only applicable for engineering metals (Poisson#s ratio 0.3) using the equivalent conical indenter of the Berkovich. However, the applicable metals are substantially limited because range of used in the finite element analysis is narrow. This study is designed to expand range of the applicable metals in the reverse algorithms established by Dao et al. and to improve the accuracy of that for extracting the elasto-plastic properties of materials. In this study, a representative strain was assumed to vary according to specific range of $E^*/{\sigma}_r$ and was defined as function of $E^*/{\sigma}_r$. Also, an initial unloading slope in reverse algorithms improved in this study was not considered as independent parameters of the load-displacement curves. The mechanical properties of materials for finite element analysis were modeled with the elastic modulus, E, the yield strength, ${\sigma}_y$, and the strain hardening exponents, n. We showed that the representative strain (0.033) suggested by Dao et al. was no longer applicable above the $E^*/{\sigma}_r$ of 400 and depended on values of $E^*/{\sigma}_r$. From these results, we constructed the dimensionless functions, in where the initial unloading slope was not included, for engineering metals up to $E^*/{\sigma}_r$ of 1500. These functions allow us to determine the mechanical properties with greater accuracy than Dao#s study.