This paper presents a novel machine learning (ML)-based method for accurate path loss prediction in cellular networks. Path loss is a critical factor in base-station positioning, and traditional methods rely on time-consuming field tests. To address this issue, the authors propose a neural network ensemble learning technique to enhance the accuracy and performance of path loss prediction. The method involves constructing an ensemble of neural networks by selecting the top-ranked networks based on hyperparameter optimization. The performance of the proposed method is compared with various ML-based methods on a public dataset, demonstrating superior accuracy and robustness compared to state-of-the-art methods. The study highlights the effectiveness of the proposed approach in predicting path loss, making it a valuable tool for improving base-station positioning in cellular networks.This paper presents a novel machine learning (ML)-based method for accurate path loss prediction in cellular networks. Path loss is a critical factor in base-station positioning, and traditional methods rely on time-consuming field tests. To address this issue, the authors propose a neural network ensemble learning technique to enhance the accuracy and performance of path loss prediction. The method involves constructing an ensemble of neural networks by selecting the top-ranked networks based on hyperparameter optimization. The performance of the proposed method is compared with various ML-based methods on a public dataset, demonstrating superior accuracy and robustness compared to state-of-the-art methods. The study highlights the effectiveness of the proposed approach in predicting path loss, making it a valuable tool for improving base-station positioning in cellular networks.