Recent years have witnessed the progress of sequential recommendation in accurately predicting users' future behaviors. However, only persuading accuracy leads to the risk of filter bubbles where recommenders only focus on users' main interest areas. Different from other studies which improves diversity or coverage, we investigate the calibration in sequential recommendation. We propose an end-to-end framework to provide both accurate and calibrated recommendations. We propose a loss function for sequential recommendation framework. In addition, we design a dual-aggregation model to further improve the recommendation. Experiments on two benchmark datasets demonstrate the effectiveness and efficiency of our model.
翻译:近些年来,在准确预测用户未来行为方面,相继建议取得了进展。然而,只有说服准确性才能导致过滤泡沫的风险,因为推荐者只关注用户的主要利益领域。不同于提高多样性或覆盖面的其他研究,我们用相继建议调查校准情况。我们提出了一个端对端框架,以提供准确和校准的建议。我们为相继建议框架提出了一个损失函数。此外,我们设计了一个双重汇总模型,以进一步改进建议。对两个基准数据集的实验显示了我们模型的有效性和效率。