Click-Conversion Multi-Task Model with Position Bias Mitigation for Sponsored Search in eCommerce
Open Access
- 18 July 2023
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 1884-1888
Abstract
Position bias, the phenomenon whereby users tend to focus on higher-ranked items of the search result list regardless of the actual relevance to queries, is prevailing in many ranking systems. Position bias in training data biases the ranking model, leading to increasingly unfair item rankings, click-through-rate (CTR), and conversion rate (CVR) predictions. To jointly mitigate position bias in both item CTR and CVR prediction, we propose two position-bias-free CTR and CVR prediction models: Position-Aware Click-Conversion (PACC) and PACC via Position Embedding (PACC-PE). PACC is built upon probability decomposition and models position information as a probability. PACC-PE utilizes neural networks to model product-specific position information as embedding. Experiments on the E-commerce sponsored product search dataset show that our proposed models have better ranking effectiveness and can greatly alleviate position bias in both CTR and CVR prediction.Keywords
All Related Versions
This publication has 19 references indexed in Scilit:
- Unbiased LambdaMART: An Unbiased Pairwise Learning-to-Rank Algorithm2019
- Estimating Position Bias without Intrusive Interventions2019
- Unbiased Learning to Rank with Unbiased Propensity Estimation2018
- Position Bias Estimation for Unbiased Learning to Rank in Personal Search2018
- Unbiased Learning-to-Rank with Biased Feedback2017
- Model Ensemble for Click Prediction in Bing Search Ads2017
- Position-normalized click prediction in search advertising2012
- An experimental comparison of click position-bias models2008
- Evaluating the accuracy of implicit feedback from clicks and query reformulations in Web searchACM Transactions on Information Systems, 2007
- Accurately interpreting clickthrough data as implicit feedback2005