This paper proposes a data-based algorithm to generate pseudo-measurements in Active Distribution Networks with a high penetration of distributed generations and a limited number of real measurements. A new technique is proposed to enhance the accuracy of pseudo-measurements. Three Deep Neural Networks (DNNs) with different hyperparameters are considered as base learners. Outputs of these DNNs are generated from available measurements and compared in terms of Mean Percentage Error. The most accurate prediction from the base learners is considered as an input for training a random forest meta learner to improve the pseudo-measurements results. Furthermore, pseudo-measurements generated from the proposed method along with the real measurements are fed into a Weighted Least Squares method to perform state estimation calculations. The effectiveness of our proposed method is evaluated using a modified IEEE standard 69 bus distribution network considering limited available measurements in presence of Distributed Generations (DGs). It is shown that the proposed method provides accurate pseudo-measurements and state estimation calculations.