The increasing penetration of low carbon technologies and electric vehicles is expected to place a considerable burden to Low Voltage (LV) power distribution networks. The previously passive management of LV networks will no longer be sustainable to cope with this growth. A more proactive approach is required to manage LV networks, usually achieved by using Power System State Estimation (PSSE) tools. The effective use of PSSE tools however, requires accurate reporting of each LV network's topology and configuration of assets (cables) that compose it. This paper presents a method for ongoing accuracy improvement of the LV networks topologies, by automating the process of approximating missing cable information and validating the network architectures. The method proposed was derived based on the assumptions that a specific network infrastructure was selected to meet a certain distributed customer loading requirement. The results obtained agrees with the assumptions, providing also a mechanism that scores the confidence level for the choices made to approximate any missing asset information.