# Multi-Feature Integration for Perception-Dependent Examination-Bias Estimation

## TL;DR

This paper introduces a Multi-Feature Integration Model (MFIM) that estimates examination bias by considering document representation and perception factors, improving bias correction in search ranking models.

## Contribution

The paper proposes a novel examination bias estimator that incorporates document features and perception factors, extending beyond traditional position-based models.

## Key findings

- MFIM outperforms existing methods in real-world experiments.
- The model demonstrates superior robustness and effectiveness.
- Mining slipoff counts enhances perception-bias modeling.

## Abstract

Eliminating examination bias accurately is pivotal to apply click-through data to train an unbiased ranking model. However, most examination-bias estimators are limited to the hypothesis of Position-Based Model (PBM), which supposes that the calculation of examination bias only depends on the rank of the document. Recently, although some works introduce information such as clicks in the same query list and contextual information when calculating the examination bias, they still do not model the impact of document representation on search engine result pages (SERPs) that seriously affects one's perception of document relevance to a query when examining. Therefore, we propose a Multi-Feature Integration Model (MFIM) where the examination bias depends on the representation of document except the rank of it. Furthermore, we mine a key factor slipoff counts that can indirectly reflects the influence of all perception-bias factors. Real world experiments on Baidu-ULTR dataset demonstrate the superior effectiveness and robustness of the new approach. The source code is available at \href{https://github.com/lixsh6/Tencent_wsdm_cup2023/tree/main/pytorch_unbias}{https://github.com/lixsh6/Tencent\_wsdm\_cup2023}