With the outbreak of today's streaming data, sequential recommendation is a promising solution to achieve time-aware personalized modeling. It aims to infer the next interacted item of given user based on history item sequence. Some recent works tend to improve the sequential recommendation via randomly masking on the history item so as to generate self-supervised signals. But such approach will indeed result in sparser item sequence and unreliable signals. Besides, the existing sequential recommendation is only user-centric, i.e., based on the historical items by chronological order to predict the probability of candidate items, which ignores whether the items from a provider can be successfully recommended. The such user-centric recommendation will make it impossible for the provider to expose their new items and result in popular bias. In this paper, we propose a novel Dual Contrastive Network (DCN) to generate ground-truth self-supervised signals for sequential recommendation by auxiliary user-sequence from item-centric perspective. Specifically, we propose dual representation contrastive learning to refine the representation learning by minimizing the euclidean distance between the representations of given user/item and history items/users of them. Before the second contrastive learning module, we perform next user prediction to to capture the trends of items preferred by certain types of users and provide personalized exploration opportunities for item providers. Finally, we further propose dual interest contrastive learning to self-supervise the dynamic interest from next item/user prediction and static interest of matching probability. Experiments on four benchmark datasets verify the effectiveness of our proposed method. Further ablation study also illustrates the boosting effect of the proposed components upon different sequential models.
翻译:随着今天流流数据的爆发,顺序建议是一个很有希望的解决方案,可以实现时间觉悟的个人化模型。它旨在根据历史项目序列推断某个用户的下一个互动项目。最近的一些工程倾向于通过随机遮掩历史项目来改进顺序建议,从而产生自我监督的信号。但这种方法确实将导致项目序列更加稀少,信号不可靠。此外,现有的顺序建议仅以用户为中心,即,根据历史项目按时间顺序顺序预测候选项目的概率,这忽略了能否成功推荐供应商的物品。这种以用户为中心的建议将使供应商无法披露其新项目并导致流行偏差。在本文件中,我们提出一个新的双重对比网络(DCN)来生成由辅助用户从项目中心角度的辅助用户后果产生的地盘自我监督信号。具体地,我们提出双重对比性学习,通过最大限度地减少给用户/项目下一个表示的兴趣之间的远差,而以用户-项目和历史项目预测的双重对比性分析部分,我们通过学习某些用户/历史项目的双重对比性分析模型,我们进一步学习它们。