Modeling user actions in job search
AF Wicaksono, A Moffat, J Zobel
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Springer | Published : 2019
© Springer Nature Switzerland AG 2019. Users of online job search tools interact with result pages in three different ways: via impressions, via clicks, and via applications. We investigate the relationship between these three kinds of interaction using logs provided by Seek.com, an Australian-based job search service. Our focus is on understanding the extent to which the three interaction types can be used to predict each other. In particular we examine models for inferring impressions from clicks, thereby providing system designers with new options for evaluating search result pages.
Related Projects (1)
Awarded by Australian Research Council
This work was supported by the Australian Research Council’s Linkage Projects scheme (project number LP150100252) and by Seek.com.