All-domain Moveline Evolution Network for Click-Through Rate Prediction

TL;DR

All-domain Moveline Evolution Network (AMEN) is proposed, which not only transfers interactions between items and scenes to homogeneous representation spaces, but also introduces a Temporal Sequential Pairwise (TSP) mechanism to understand the nuanced associations between scene-level and item-level behaviors.

Abstract

E-commerce app users exhibit behaviors that are inherently logically consistent. A series of multi-scenario user behaviors interconnect to form the scene-level all-domain user moveline, which ultimately reveals the user's true intention. Traditional CTR prediction methods typically focus on the item-level interaction between the target item and the historically interacted items. However, the scene-level interaction between the target item and the user moveline remains underexplored. There are two challenges when modeling the interaction with preceding all-domain user moveline: (i) Heterogeneity between items and scenes: Unlike traditional user behavior sequences that utilize items as carriers, the user moveline utilizes scenes as carriers. The heterogeneity between items and scenes complicates the process of aligning interactions within a unified representation space. (ii) Temporal misalignment of linked scene-level and item-level behaviors: In the preceding user moveline with a fixed sampling length, certain critical scene-level behaviors are closely linked to subsequent item-level behaviors. However, it is impossible to establish a complete temporal alignment that clearly identifies which specific scene-level behaviors correspond to which item-level behaviors. To address these challenges and pioneer modeling user intent from the perspective of the all-domain moveline, we propose All-domain Moveline Evolution Network (AMEN). AMEN not only transfers interactions between items and scenes to homogeneous representation spaces, but also introduces a Temporal Sequential Pairwise (TSP) mechanism to understand the nuanced associations between scene-level and item-level behaviors, ensuring that the all-domain user moveline differentially influences CTR predictions for user's favored and unfavored items. Online A/B testing demonstrates that our method achieves a +11.6% increase in CTCVR.

Figures (4)

• Figure 1: A typical showcase of how the preceding scene-level all-domain user moveline uncovers user's true intention. A user visits the app's marketing channel twice at time $t_0$ and $t_1$, with preceding user moveline $\mathcal{\vec{M}}_{t_0}$ and $\mathcal{\vec{M}}_{t_1}$, respectively.
• Figure 2: The architecture of All-domain Moveline Evolution Network (AMEN). The CTR prediction module serves as the online inference architecture, which merges the outputs of item-to-scene Scene-level Attention and scene-to-item Item-level Attention. The TSP mechanism in the Scene-level Attention ensures that the preceding all-domain user moveline distinctly impacts user's favored and unfavored items.
• Figure 3: Item-to-Scene in the inference feed-forward stage, generating the moveline reward $r$.
• Figure 4: Probability density distribution of the moveline reward.