FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W17303057> ?p ?o ?g. }

• W17303057 startingPage "16" @default.
• W17303057 abstract "ABSTRACT We study a comprehensive framework for sponsored search which incorporates advertiser budgets, query frequency forecasts, and pricing and ranking schemes. We propose a linear program for optimizing revenue (or the total value to advertisers) that has an exponential number of variables; however, we describe how it can be solved efficiently using column generation. The formulation is easily extendable to various levels of problem complexity, adaptable to dynamic environments, fast, and works well in terms of practical considerations. Simulations show significant improvements in revenue and efficiency. Keywords: column generation, sponsored search, budgets, advertising (ProQuest: ... denotes formulae omitted.) 1. Introduction Electronic commerce is thriving [Kim 2004; Chen 2003; Alczak 2006] and one of the driving forces behind its success is internet advertising. Search engine companies earn millions of dollars every day by auctioning off advertisement slots. Internet advertising is primarily accomplished through Sponsored Search, ads placed on publisher pages and banner ads. In this paper we will primarily be concerned with Sponsored Search, but we note that scheduling of ads for third party publishers can benefit from a very similar treatment. In Sponsored Search there are several sources of data which must be considered. Firstly, there are the actual bids by advertisers for the search terms (queries), secondly the distribution of query frequencies and finally the advertiser budgets, when these are declared. The query frequencies limit the number of times a search engine can display its advertisers. Note that, unlike the bids and budgets, query frequencies are not under the control of the advertisers or, for that matter, of the search engine. It is well known [Silverstein 1999] that search engine query frequency distribution typically has relatively few queries with large volume and revenue, and a very large number of queries with extremely low volume. Therefore, we overcome most of the uncertainty in query volumes by selecting a relatively small subset of queries whose near-term volumes are easy to forecast, yet still constitute a large amount of the overall revenue. Advertisers or their agents, on the other hand, do have the ability to control their budgets. An advertiser's budget may constrain the number of times their ads appear, even when they have made a high bid on a query term. One might ask why they would wish to do so. There are several possible reasons, among them: protection against click-fraud, an over-all company advertising budget, and the desire to control the allocation of that budget between various media and campaigns. Whatever the reason, the search engine must determine which advertisers to display for which queries, given these constraints. Pricing and ranking are additional parameters of the system that influence revenue. The VCG mechanism [Vickrey 1961; Clarke 1971; Groves 1973] can be applied, but in practice search engines predominantly use the generalized second price (GSP) auction, described in more detail by Edelman et al. [Edelman 2006]. Advertisers are ranked according to the product of the price they bid for receiving a click, and a quality score. Each is charged a price per click equal to the minimum they would have had to pay to maintain their rank. The problem we consider is how to allocate advertisers to queries such that budget constraints are satisfied and efficiency or revenue is maximized. This problem is posed as a linear program that takes a global view, and coordinates advertiser spend across the chosen time-period, such as the next hour, or an entire day. With the combined knowledge of forecast query volumes, advertiser budgets, advertiser bids and the pricing and ranking algorithm, we formulate a comprehensive mathematical framework. Because of the flexibility of LP models, we are able to extend our approach beyond its original form to take onto consideration multi-level budgets, dynamic changes in the data, and a number of other practical extensions. …" @default.
• W17303057 created "2016-06-24" @default.
• W17303057 creator A5004541836 @default.
• W17303057 creator A5009442286 @default.
• W17303057 creator A5030790002 @default.
• W17303057 creator A5056835997 @default.
• W17303057 date "2008-02-01" @default.
• W17303057 modified "2023-09-24" @default.
• W17303057 title "Ad Delivery with Budgeted Advertisers: A Comprehensive LP Approach" @default.
• W17303057 cites W1678923740 @default.
• W17303057 cites W1769120835 @default.
• W17303057 cites W1970355999 @default.
• W17303057 cites W1982889956 @default.
• W17303057 cites W1986345564 @default.
• W17303057 cites W2015007620 @default.
• W17303057 cites W2015324401 @default.
• W17303057 cites W2038701206 @default.
• W17303057 cites W2078040677 @default.
• W17303057 cites W2114632363 @default.
• W17303057 cites W2120100612 @default.
• W17303057 cites W2127347613 @default.
• W17303057 cites W2127470768 @default.
• W17303057 cites W2135724610 @default.
• W17303057 cites W2158701722 @default.
• W17303057 cites W2910445340 @default.
• W17303057 cites W3122305203 @default.
• W17303057 cites W38064460 @default.
• W17303057 cites W72737698 @default.
• W17303057 cites W1772070704 @default.
• W17303057 hasPublicationYear "2008" @default.
• W17303057 type Work @default.
• W17303057 sameAs 17303057 @default.
• W17303057 citedByCount "9" @default.
• W17303057 countsByYear W173030572014 @default.
• W17303057 countsByYear W173030572016 @default.
• W17303057 countsByYear W173030572019 @default.
• W17303057 countsByYear W173030572021 @default.
• W17303057 crossrefType "journal-article" @default.
• W17303057 hasAuthorship W17303057A5004541836 @default.
• W17303057 hasAuthorship W17303057A5009442286 @default.
• W17303057 hasAuthorship W17303057A5030790002 @default.
• W17303057 hasAuthorship W17303057A5056835997 @default.
• W17303057 hasConcept C110875604 @default.
• W17303057 hasConcept C112698675 @default.
• W17303057 hasConcept C121955636 @default.
• W17303057 hasConcept C136764020 @default.
• W17303057 hasConcept C144133560 @default.
• W17303057 hasConcept C162324750 @default.
• W17303057 hasConcept C175444787 @default.
• W17303057 hasConcept C187687199 @default.
• W17303057 hasConcept C189430467 @default.
• W17303057 hasConcept C195487862 @default.
• W17303057 hasConcept C21782646 @default.
• W17303057 hasConcept C23123220 @default.
• W17303057 hasConcept C33923547 @default.
• W17303057 hasConcept C41008148 @default.
• W17303057 hasConcept C42475967 @default.
• W17303057 hasConcept C512338625 @default.
• W17303057 hasConcept C97854310 @default.
• W17303057 hasConceptScore W17303057C110875604 @default.
• W17303057 hasConceptScore W17303057C112698675 @default.
• W17303057 hasConceptScore W17303057C121955636 @default.
• W17303057 hasConceptScore W17303057C136764020 @default.
• W17303057 hasConceptScore W17303057C144133560 @default.
• W17303057 hasConceptScore W17303057C162324750 @default.
• W17303057 hasConceptScore W17303057C175444787 @default.
• W17303057 hasConceptScore W17303057C187687199 @default.
• W17303057 hasConceptScore W17303057C189430467 @default.
• W17303057 hasConceptScore W17303057C195487862 @default.
• W17303057 hasConceptScore W17303057C21782646 @default.
• W17303057 hasConceptScore W17303057C23123220 @default.
• W17303057 hasConceptScore W17303057C33923547 @default.
• W17303057 hasConceptScore W17303057C41008148 @default.
• W17303057 hasConceptScore W17303057C42475967 @default.
• W17303057 hasConceptScore W17303057C512338625 @default.
• W17303057 hasConceptScore W17303057C97854310 @default.
• W17303057 hasIssue "1" @default.
• W17303057 hasLocation W173030571 @default.
• W17303057 hasOpenAccess W17303057 @default.
• W17303057 hasPrimaryLocation W173030571 @default.
• W17303057 hasRelatedWork W1483499855 @default.
• W17303057 hasRelatedWork W18167294 @default.
• W17303057 hasRelatedWork W1881046052 @default.
• W17303057 hasRelatedWork W1937230960 @default.
• W17303057 hasRelatedWork W1938563015 @default.
• W17303057 hasRelatedWork W1996312099 @default.
• W17303057 hasRelatedWork W2038263585 @default.
• W17303057 hasRelatedWork W2072006785 @default.
• W17303057 hasRelatedWork W2117340018 @default.
• W17303057 hasRelatedWork W2119914577 @default.
• W17303057 hasRelatedWork W2131951207 @default.
• W17303057 hasRelatedWork W2135724610 @default.
• W17303057 hasRelatedWork W2138559067 @default.
• W17303057 hasRelatedWork W2254816 @default.
• W17303057 hasRelatedWork W2369562334 @default.
• W17303057 hasRelatedWork W2585510658 @default.
• W17303057 hasRelatedWork W2732901914 @default.
• W17303057 hasRelatedWork W2995289083 @default.
• W17303057 hasRelatedWork W3122305203 @default.

• next