User-aware advertisability

Hai Tao Yu, Tetsuya Sakai

Research output: Chapter in Book/Report/Conference proceedingConference contribution


In sponsored search, many studies focus on finding the most relevant advertisements (ads) and their optimal ranking for a submitted query. Determining whether it is suitable to show ads has received less attention. In this paper, we introduce the concept of user-aware advertisability, which refers to the probability of ad-click on sponsored ads when a specific user submits a query. When computing the advertisability for a given query-user pair, we first classify the clicked web pages based on a pre-defined category hierarchy and use the aggregated topical categories of clicked web pages to represent user preference. Taking user preference into account, we then compute the ad-click probability for this query-user pair. Compared with existing methods, the experimental results show that user preference is of great value for generating user-specific advertisability. In particular, our approach that computes advertisability per query-user pair outperforms the two state-of-the-art methods that compute advertisability per query in terms of a variant of the normalized Discounted Cumulative Gain metric.

Original languageEnglish
Title of host publicationInformation Retrieval Technology - 9th Asia Information Retrieval Societies Conference, AIRS 2013, Proceedings
Number of pages12
Publication statusPublished - 2013
Event9th Asia Information Retrieval Societies Conference on Information Retrieval Technology, AIRS 2013 - Singapore, Singapore
Duration: 2013 Dec 92013 Dec 11

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume8281 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Conference9th Asia Information Retrieval Societies Conference on Information Retrieval Technology, AIRS 2013

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)


Dive into the research topics of 'User-aware advertisability'. Together they form a unique fingerprint.

Cite this