Hard Example Mining based Adversarial Autoencoder Recommendation Algorithm

Commonly used datasets in recommendation research suffer from unbalanced data distribution, sparsity, and different user rating preferences. All these problems affect the quality of recommendation. Thus, this paper proposed a recommendation model by combining hard example mining with adversarial autoencoder. Considering the difference in users' preference, Mean Model based triplet loss algorithm was introduced to classify the dataset into positive and negative samples and thus improve the quality of the training data. Using classified samples, the rating prediction model was trained from both reconstruction and adversarial aspects. Adam optimization algorithm was used to calculate different update gradients for different parameters. Experimental results show that the recommendation model improves the recommendation accuracy significantly, and several performance indicators are better than baseline models.