Research on the Prediction Model of Automobile Digital Marketing Effect Based on Deep Learning

This study aims to develop an advanced prediction model that uses deep learning algorithms to estimate the effectiveness of automotive digital marketing activities. By combining the advantages of MLP, CNN and LSTM, a composite model is constructed to improve the accuracy of the prediction model. This study first constructs a systematic feature extraction and data preprocessing process, and uses the MLP algorithm to establish a preliminary framework of the prediction model. Furthermore, combining the powerful ability of CNN in feature recognition and the expertise of LSTM in processing time series data, an in-depth prediction analysis of the marketing effect is performed. Through a series of experiments and model simulation verification, the study found that the proposed hybrid model shows high accuracy in multidimensional data prediction tasks. The simulation results not only provide accurate data analysis, but also confirm the applicability and advantages of the model in a changing marketing environment. Finally, through comparative analysis with traditional prediction methods, this paper further verifies the effectiveness and broad application prospects of deep learning technology in the field of automotive digital marketing effect prediction.