Middleman Bias in Advertising: Aligning Relevance of Keyphrase Recommendations with Search
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866908381025927168 |
|---|---|
| author | Dey, Soumik Zhang, Wei Wu, Hansi Dong, Bingfeng Li, Binbin |
| author_facet | Dey, Soumik Zhang, Wei Wu, Hansi Dong, Bingfeng Li, Binbin |
| contents | E-commerce sellers are recommended keyphrases based on their inventory on which they advertise to increase buyer engagement (clicks/sales). Keyphrases must be pertinent to items; otherwise, it can result in seller dissatisfaction and poor targeting -- towards that end relevance filters are employed. In this work, we describe the shortcomings of training relevance filter models on biased click/sales signals. We re-conceptualize advertiser keyphrase relevance as interaction between two dynamical systems -- Advertising which produces the keyphrases and Search which acts as a middleman to reach buyers. We discuss the bias of search relevance systems (middleman bias) and the need to align advertiser keyphrases with search relevance signals. We also compare the performance of cross encoders and bi-encoders in modeling this alignment and the scalability of such a solution for sellers at eBay. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_00131 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Middleman Bias in Advertising: Aligning Relevance of Keyphrase Recommendations with Search Dey, Soumik Zhang, Wei Wu, Hansi Dong, Bingfeng Li, Binbin Information Retrieval E-commerce sellers are recommended keyphrases based on their inventory on which they advertise to increase buyer engagement (clicks/sales). Keyphrases must be pertinent to items; otherwise, it can result in seller dissatisfaction and poor targeting -- towards that end relevance filters are employed. In this work, we describe the shortcomings of training relevance filter models on biased click/sales signals. We re-conceptualize advertiser keyphrase relevance as interaction between two dynamical systems -- Advertising which produces the keyphrases and Search which acts as a middleman to reach buyers. We discuss the bias of search relevance systems (middleman bias) and the need to align advertiser keyphrases with search relevance signals. We also compare the performance of cross encoders and bi-encoders in modeling this alignment and the scalability of such a solution for sellers at eBay. |
| title | Middleman Bias in Advertising: Aligning Relevance of Keyphrase Recommendations with Search |
| topic | Information Retrieval |
| url | https://arxiv.org/abs/2502.00131 |