Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

MetaSplit: Meta-Split Network for Limited-Stock Product Recommendation

Wenhao Wu, Jialiang Zhou, Ailong He, Shuguang Han, Jufeng Chen, Bo Zheng

TL;DR

The Meta-Split Network (MSNet) is proposed to split user history sequence regarding to the volume of stock for each product, and adopt differentiated modeling approaches for different sequences to address the problem of inconvergence.

Abstract

Compared to business-to-consumer (B2C) e-commerce systems, consumer-to-consumer (C2C) e-commerce platforms usually encounter the limited-stock problem, that is, a product can only be sold one time in a C2C system. This poses several unique challenges for click-through rate (CTR) prediction. Due to limited user interactions for each product (i.e. item), the corresponding item embedding in the CTR model may not easily converge. This makes the conventional sequence modeling based approaches cannot effectively utilize user history information since historical user behaviors contain a mixture of items with different volume of stocks. Particularly, the attention mechanism in a sequence model tends to assign higher score to products with more accumulated user interactions, making limited-stock products being ignored and contribute less to the final output. To this end, we propose the Meta-Split Network (MSNet) to split user history sequence regarding to the volume of stock for each product, and adopt differentiated modeling approaches for different sequences. As for the limited-stock products, a meta-learning approach is applied to address the problem of inconvergence, which is achieved by designing meta scaling and shifting networks with ID and side information. In addition, traditional approach can hardly update item embedding once the product is consumed. Thereby, we propose an auxiliary loss that makes the parameters updatable even when the product is no longer in distribution. To the best of our knowledge, this is the first solution addressing the recommendation of limited-stock product. Experimental results on the production dataset and online A/B testing demonstrate the effectiveness of our proposed method.

MetaSplit: Meta-Split Network for Limited-Stock Product Recommendation

TL;DR

The Meta-Split Network (MSNet) is proposed to split user history sequence regarding to the volume of stock for each product, and adopt differentiated modeling approaches for different sequences to address the problem of inconvergence.

Abstract

Compared to business-to-consumer (B2C) e-commerce systems, consumer-to-consumer (C2C) e-commerce platforms usually encounter the limited-stock problem, that is, a product can only be sold one time in a C2C system. This poses several unique challenges for click-through rate (CTR) prediction. Due to limited user interactions for each product (i.e. item), the corresponding item embedding in the CTR model may not easily converge. This makes the conventional sequence modeling based approaches cannot effectively utilize user history information since historical user behaviors contain a mixture of items with different volume of stocks. Particularly, the attention mechanism in a sequence model tends to assign higher score to products with more accumulated user interactions, making limited-stock products being ignored and contribute less to the final output. To this end, we propose the Meta-Split Network (MSNet) to split user history sequence regarding to the volume of stock for each product, and adopt differentiated modeling approaches for different sequences. As for the limited-stock products, a meta-learning approach is applied to address the problem of inconvergence, which is achieved by designing meta scaling and shifting networks with ID and side information. In addition, traditional approach can hardly update item embedding once the product is consumed. Thereby, we propose an auxiliary loss that makes the parameters updatable even when the product is no longer in distribution. To the best of our knowledge, this is the first solution addressing the recommendation of limited-stock product. Experimental results on the production dataset and online A/B testing demonstrate the effectiveness of our proposed method.
Paper Structure (23 sections, 22 equations, 3 figures, 4 tables)

This paper contains 23 sections, 22 equations, 3 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Cold-Start Recommendation
  4. Meta Learning for Recommendation
  5. Preliminaries
  6. Problem Formulation
  7. Production CTR Prediction Model
  8. Methodology
  9. Architecture Overview
  10. Detail Structures
  11. Sequence Split Module
  12. Sequence Meta Learning Module
  13. Auxiliary Loss
  14. Experiments
  15. Experiment Setup
  16. ...and 8 more sections

Figures (3)

  • Figure 1: An illustrative example highlighting the distinctions between B2C platforms and C2C platforms.
  • Figure 2: An illustration of Embedding and MLP structure with sequntial information modeling for the deep CTR prediction model
  • Figure 3: The schematic framework of MSNet can be broadly divided into three components: the sequence split module, the sequence meta-learning module, and the auxiliary loss. Figure (a) provides an overview of the model, with the sequence split module depicted at the bottom. Figure (b) illustrates the sequence meta-learning module, while Figure (c) offers detailed insights into the auxiliary loss.