Distributed Web Search as a Stochastic Game

Khoussainov, Rinat; Kushmerick, Nicholas

doi:10.1007/978-3-540-24610-7_5

Cited by 2 publications

(6 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus, for the first two cases (when gi(·; S, t) is strictly convex), Inequality (5) is strict for this (S, t) pair. 10 Again, the reason for the strictness is slightly different in the third case. Focus on the searcher types (S, t) ∈ X ∪ X .…”

Section: Discussionmentioning

confidence: 90%

See 1 more Smart Citation

User satisfaction in competitive sponsored search

Kempe

Lucier

2014

Proceedings of the 23rd International Conference on World Wide Web

View full text Add to dashboard Cite

We present a model of competition between web search algorithms, and study the impact of such competition on user welfare. In our model, search providers compete for customers by strategically selecting which search results to display in response to user queries. Customers, in turn, have private preferences over search results and will tend to use search engines that are more likely to display pages satisfying their demands.Our main question is whether competition between search engines increases the overall welfare of the users (i.e., the likelihood that a user finds a page of interest). When search engines derive utility only from customers to whom they show relevant results, we show that they differentiate their results, and every equilibrium of the resulting game achieves at least half of the welfare that could be obtained by a social planner. This bound also applies whenever the likelihood of selecting a given engine is a convex function of the probability that a user's demand will be satisfied, which includes natural Markovian models of user behavior.On the other hand, when search engines derive utility from all customers (independent of search result relevance) and the customer demand functions are not convex, there are instances in which the (unique) equilibrium involves no differentiation between engines and a high degree of randomness in search results. This can degrade social welfare by a factor of Ω( √ N ) relative to the social optimum, where N is the number of webpages. These bad equilibria persist even when search engines can extract only small (but non-zero) expected revenue from dissatisfied users, and much higher revenue from satisfied ones.

show abstract

Section: Discussionmentioning

confidence: 90%

“…A somewhat similar model of search engine competition was proposed by Khoussainov and Kushmerick [9,10]. They consider a metasearcher that directs queries to specialized engines, and propose and experimentally evaluate reinforcement learning approaches for the search engines.…”

Section: Related Workmentioning

confidence: 98%

User satisfaction in competitive sponsored search

Kempe

Lucier

2014

Proceedings of the 23rd International Conference on World Wide Web

View full text Add to dashboard Cite

show abstract

“…A somewhat similar model of search engine competition was proposed by Khoussainov and Kushmerick [10,11]. They consider a metasearcher that directs queries to specialized engines.…”

Section: Related Workmentioning

confidence: 99%

“…The selection rule used in their model is the Majority Selection Rule from Section 2.4 of our paper, and a search engine derives full benefit from all users that are directed to it (in our terminology, the case β = 1). Rather than studying the equilibria of the resulting game, Khoussainov and Kushmerick [10,11] propose and experimentally evaluate reinforcement learning approaches for the search engines.…”

Section: Related Workmentioning

confidence: 99%

“…For λ in the range (zγ(n * ), ∞), we have γ(n) λ−zγ(n) > 0 for all pages n, and the expression n γ(n) λ−zγ(n) is strictly decreasing, approaching 0 as λ → ∞ and approaching ∞ as λ → zγ(n * ). There is therefore a unique solution (in λ ′ i ) to the Equation (11) in the range (zγ(n * ), ∞). By choice of n * , all terms of n γ(n) λ ′ i −zγ(n) are positive at this solution, and each γ(n) λ ′ i −zγ(n) lies in (0, 1).…”

Section: Existence and Uniqueness Of Symmetric Equilibriamentioning

confidence: 99%

See 1 more Smart Citation