Improved CTR Prediction Algorithm based on LSTM and Attention
Qiaojun Chen, Dong Li · 2021
Click-through rate prediction is a core research issue in recommendation systems and online advertising, playing a pivotal role in the entire Internet field. This article mainly integrates the LSTM module and the attention module into the DeepFM framework, and proposes LDeepFM and LADeepFM which can extract time series information more effectively. We use real data sets to evaluate the prediction performance of the model. The experiment results demonstrate that our proposed approach can increase the accuracy of click-through rate prediction more effectively than traditional models.