An improved hybrid semi-stacked autoencoder for item-features of recommendation system (iHSARS)

In recent years, information overload has become a phenomenon where it makes people difficult to filter relevant information. To address issues such as high-dimensional data, cold start, and data sparsity, semi-autoencoder is one of the unsupervised deep learning methods used in recommendation systems. It particularly useful for reducing dimensions, capturing latent representations, flexibly reconstructing various parts input data. this article, we propose an improved hybrid semi-stacked autoencoder item-features system (iHSARS) framework. This method aims show better performance collaborative via (HRSA) technique. Two novel elements iHSARS’s architecture have been introduced. The first element increase sources side layer, while second number hidden layers expanded. verify improvement model, MovieLens-100K MovieLens-1M datasets applied model. comparison between proposed model different state-of-the-art carried using mean absolute error (MAE) root square (RMSE) metrics. experiments demonstrate that our framework efficiency than others.