FE-TCM: Filter-Enhanced Transformer Click Model for Web Search

文献类型： 外文期刊

第一作者： Wang, Yingfei

作者： Wang, Yingfei;Liu, Jianping;Wang, Xiaofeng;Wang, Meng;Chu, Xintao;Liu, Jianping;Wang, Jian

作者机构：

关键词： Transformers; Behavioral sciences; Predictive models; Data models; Search engines; Discrete Fourier transforms; Information filters; Click model; click prediction; web search; transformer

期刊名称：IEEE ACCESS （ 影响因子：3.9； 五年影响因子：4.1 ）

ISSN： 2169-3536

年卷期： 2023 年 11 卷

页码：

收录情况： SCI

摘要： Constructing click models and extracting implicit relevance feedback information from interaction between users and search engines are very important for improving the ranking of search results. Neural networks are effective for modeling users' click behavior, and we propose a novel Filter-Enhanced Transformer Click Model (FE-TCM) for web search. The model uses the powerful Transformer model as the backbone network for feature extraction and innovatively add a filter layer. Firstly, in order to reduce the influence of noise on user behavior data, we use the learnable filters to filter the log noise. Secondly, following the examination hypothesis, we model the attraction estimator and examination predictor respectively to output attractiveness scores and examination probabilities. A novel transformer model is used to learn the deeper representation among different features. Finally, we apply the different combination functions to integrate attractiveness scores and examination probabilities into the click prediction. From our experiments on two real-world session datasets, it is proved that FE-TCM outperforms the existing click models for the click prediction.

分类号：

  • 相关文献
作者其他论文 更多>>

微信小程序