Abstract

Abstract:
Sponsored search is one of the enabling technologies for today's Web searchengines. It corresponds to matching and showing ads related to theuser query on the search engine results page. Users are likely toclick on topically related ads and the advertisers pay only when auser clicks on their ad. Hence, it isimportant to be able to predict if an ad is likely to be clicked, and maximizethe number of clicks. We investigate the sponsored search problem from a machinelearning perspective with respect to three main sub-problems: how to use clickdata for training and evaluation, which learning framework is moresuitable for the task, and which features are useful for existingmodels. We perform a large scale evaluation based on datafrom a commercial Web search engine. Results show that it ispossible to learn and evaluate directly and exclusively on click dataencoding pairwise preferences following simple and conservativeassumptions. Furthermore, we find that online multilayer perceptron learning, basedon a small set of features representing content similarity ofdifferent kinds, significantly outperforms an information retrievalbaseline and other learning models, providing a suitable framework forthe sponsored search task.


Download:

P227 ciaramita 0.pdf
ACM COPYRIGHT NOTICE. Copyright © 2012 by the Association for Computing Machinery, Inc. Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, to republish, to post on servers, or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from Publications Dept., ACM, Inc., fax +1 (212) 869-0481, or permissions@acm.org.