Style4Rec: Enhancing Transformer-based E-commerce Recommendation Systems with Style and Shopping Cart Information

TL;DR

Style4Rec addresses the gap in transformer-based sequential recommendations by incorporating image style information and shopping cart data. It introduces a two-part model with a deep transformer encoder and a neural style transfer–based style embedding module, augmented by a training strategy that distinguishes purchase from shopping cart sessions. Empirical results show consistent improvements over BERT4Rec and SASRec across HR, NDCG, and MRR at multiple cutoffs, with ablations confirming the contributions of style embeddings and cart data. The approach demonstrates a scalable pathway to richer user preference modeling in e-commerce by leveraging visual style cues and cart signals to enhance personalized recommendations.

Abstract

Understanding users' product preferences is essential to the efficacy of a recommendation system. Precision marketing leverages users' historical data to discern these preferences and recommends products that align with them. However, recent browsing and purchase records might better reflect current purchasing inclinations. Transformer-based recommendation systems have made strides in sequential recommendation tasks, but they often fall short in utilizing product image style information and shopping cart data effectively. In light of this, we propose Style4Rec, a transformer-based e-commerce recommendation system that harnesses style and shopping cart information to enhance existing transformer-based sequential product recommendation systems. Style4Rec represents a significant step forward in personalized e-commerce recommendations, outperforming benchmarks across various evaluation metrics. Style4Rec resulted in notable improvements: HR@5 increased from 0.681 to 0.735, NDCG@5 increased from 0.594 to 0.674, and MRR@5 increased from 0.559 to 0.654. We tested our model using an e-commerce dataset from our partnering company and found that it exceeded established transformer-based sequential recommendation benchmarks across various evaluation metrics. Thus, Style4Rec presents a significant step forward in personalized e-commerce recommendation systems.

Figures (5)

• Figure 1: Neural style transfer algorithm updates the content image, with the help of gram matrices, depending on which layers' gram matrices are used and the ratio of $\frac{L_{style}}{L_{content}}$, different final results can be obtained.
• Figure 2: First method of neural style transfer algorithm, algorithm updates the noisy input image with the help of gram matrices.
• Figure 3: Model Architecture
• Figure 4: An example of how the repeated final products were removed, each number represents a product.
• Figure 5: The effect of maximum session length.