🤖 AI Summary
In e-commerce recommendation, the “Click Item A, Buy Item B” (CABB) phenomenon introduces attribution bias in conversion modeling, violating the conventional click-conversion alignment assumption. Method: We propose a category-aware multi-task attribution modeling framework that (i) constructs a category similarity matrix to capture substitutional and complementary relationships among items, (ii) designs a collaborative filtering–based weighting mechanism to differentiate intra-category and cross-category conversion behaviors, and (iii) jointly optimizes click-through rate (CTR) and conversion rate (CVR) prediction via multi-task learning. Contribution/Results: The approach effectively mitigates model bias stemming from rigid click-conversion alignment. Offline evaluation shows a 13.9% reduction in normalized entropy; online A/B testing demonstrates a 0.25 percentage-point improvement in core conversion rate, validating both practical effectiveness and production deployability.
📝 Abstract
User journeys in e-commerce routinely violate the one-to-one assumption that a clicked item on an advertising platform is the same item later purchased on the merchant's website/app. For a significant number of converting sessions on our platform, users click product A but buy product B -- the Click A, Buy B (CABB) phenomenon. Training recommendation models on raw click-conversion pairs therefore rewards items that merely correlate with purchases, leading to biased learning and sub-optimal conversion rates. We reframe conversion prediction as a multi-task problem with separate heads for Click A Buy A (CABA) and Click A Buy B (CABB). To isolate informative CABB conversions from unrelated CABB conversions, we introduce a taxonomy-aware collaborative filtering weighting scheme where each product is first mapped to a leaf node in a product taxonomy, and a category-to-category similarity matrix is learned from large-scale co-engagement logs. This weighting amplifies pairs that reflect genuine substitutable or complementary relations while down-weighting coincidental cross-category purchases. Offline evaluation on e-commerce sessions reduces normalized entropy by 13.9% versus a last-click attribution baseline. An online A/B test on live traffic shows +0.25% gains in the primary business metric.