Abstract: Sequential recommendation is a recommendation paradigm that models the user's historical interaction sequence to predict the user's preference for the next item. Due to the diversity and complexity of user interaction history, interaction sequences often contain noisy and irrelevant noise terms, which affects the model to capture stable sequence patterns. At the same time, there are various sequence signals in the interactive sequence, such as trend signal, semantic signal and residual signal, which cannot be accurately expressed by a single type of modeling. To solve these two problems, this paper proposes a sequential recommendation model (DeepEMA) based on deep moving average. In order to solve the noise problem in user interaction sequences, DeepEMA uses the exponential moving average method in the field of time series to smooth the sequences, filter the noise and initially extract the sequence trend. In order to capture complex and diverse sequence signals in interaction sequences, DeepEMA models the sequential trend signal, dimensional semantic signal, and residual signal respectively through a multi-module architecture including trend module, dimensional semantic module, and residual extraction module. The experimental results on four publicly available datasets demonstrate that, our proposed model demonstrates superior recommendation accuracy in terms of Hit Ratio (HR) and Normalized Discounted Cumulative Gain (NDCG) metrics. For example, there is an average improvement of 5.67% on the HR@5 metric and 4.88% on the NDCG@5 metric compared to the most advanced baseline. And the sequence noise problem is significantly improved. The ablation experiments also verified the effectiveness of each module.

Key words: Sequential recommendation, Moving average, Noise, Sequential signal

摘要： 序列推荐是对用户历史交互序列进行建模，以预测用户所偏好的下一项目的推荐范式。由于用户交互历史的多样性和复杂性，交互序列常包含嘈杂和不相关的噪声项，从而影响模型捕捉稳定的序列模式；同时，交互序列中存在多样的序列信号，例如趋势信号、语义信号和残差信号等，依赖单一模式进行建模无法得到精确的表达。针对上述两个问题，本文提出了一种基于深度化移动平均的序列推荐模型（DeepEMA）。为了解决用户交互序列中的噪音问题，DeepEMA使用时间序列领域的指数移动平均法对序列进行平滑处理，过滤噪音并初步提取序列趋势；为了捕捉交互序列中复杂多样的序列信号，DeepEMA通过包括序列趋势模块、维度语义模块及残差提取模块在内的多模块架构，分别建模序列的趋势信号、维度的语义信号以及残差信号。在4个公开的数据集上进行了验证，实验结果表明，本模型的推荐准确性在命中率（Hit Ratio, HR）和归一化折损累计增益（Normalized Discounted Cumulative Gain, NDCG）指标上均优于最先进的基线算法，例如在HR@5指标上，比最先进的基线平均提高了5.67%；在NDCG@5指标上，比最先进的基线平均提高了4.88%。并且明显的改善了序列的噪音问题，消融实验也验证了各模块的有效性。

关键词: 序列推荐, 移动平均, 噪音, 序列信号