FCNN+: An Improved Fully Connected Neural Network High-accuracy Prediction Model

The rapid development of mobile communication technology has brought convenience to people, and improving customer satisfaction has become an important indicator of the market performance of major operators. Nowadays, traditional methods are often ineffective at enhancing customer satisfaction. Therefore, we aim to improve customer satisfaction by focusing on factors that affect their online experience. This paper proposes an Improved Fully Connected Neural Network (FCNN+) to predict customer satisfaction. The final prediction evaluation metric, mean absolute error (MAE), shows that the FCNN+ framework we propose achieves high accuracy.