William Culhane
K. R. Jayaram
ABSTRACT
Top-k matching is a fundamental problem underlying on-line advertising platforms, mobile social networks, etc. Distributed processes (e.g., advertisers) specify predicates, which we call subscriptions, for events (e.g., user actions) they wish to react to. Subscriptions define weights for elementary constraints on individual event attributes and do not require that events match all constraints. An event is multicast only to the processes with the k highest match scores for that event -- this score is the aggregation of the weights of all constraints in a subscription matching the event.
However, state-of-the-art approaches to top-k matching support only rigid models of events and subscriptions, which leads to suboptimal matches. We present a novel model of weighted top-k matching which is more expressive than the state-of-the-art, and a corresponding efficient algorithm. Our model supports attributes with intervals, weights specified by producers of events or by subscriptions, negative weights, prorating of matched constraints, and the ability to vary scores dynamically with system parameters. Our algorithm exhibits time and space complexities which are competitive with state-of-the-art algorithms regardless of our added expressiveness -- O(M log N + S log k) and O(M N + k) respectively, with N the number of constraints, M the number of event attributes, and S the number of matching constraints.
Through empirical evaluation with both statistically generated and real-world data we demonstrate that our algorithm is (a) equally or more efficient and scalable than the state-of-the art without exploiting our added expressiveness, and it (b) significantly outperforms existing approaches upgraded -- if possible at all -- to match our expressiveness.
- M. Aguilera, R. Strom, D. Sturman, M. Astley, and T. Chandra. Matching Events in a Content-based Subscription System. In PODC, 1999. Google Scholar
Digital Library
- L. Arge and J. Vitter. Optimal Dynamic Interval Management in External Memory. In FOCS, 1996. Google ScholarDigital Library
- A. Bhalgat, J. Feldman, and V. Mirrokni. Online Allocation of Display Ads with Smooth Delivery. In KDD, 2012. Google ScholarDigital Library
- P. Cao and Z. Wang. Efficient Top-K Query Calculation in Distributed Networks. In PODC, 2004. Google ScholarDigital Library
- S.-C. Chu. Viral Advertising in Social Media: Participation in Facebook Groups and Responses among College-aged Users. Journal of Interactive Advertising, 12(1):30--43, 2011.Google ScholarCross Ref
- T. Cormen, R. Rivest, C. Leiserson, and C. Stein. Introduction to Algorithms. MIT Press, 2011. Google ScholarDigital Library
- W. Culhane, K. Kogan, C. Jayalath, and P. Eugster. LOOM: Optimal Aggregation Overlays for In-Memory Big Data Processing. In HotCloud, 2014. Google ScholarDigital Library
- M. Dylla, I. Miliaraki, and M. Theobald. Top-k Query Processing in Probabilistic Databases with Non-materialized Views. In ICDE, 2013. Google ScholarDigital Library
- R. Fagin. Combining Fuzzy Information from Multiple Systems. In PODS, 1996. Google ScholarDigital Library
- R. Fagin, A. Lotem, and M. Naor. Optimal Aggregation Algorithms for Middleware. In PODS, 2001. Google ScholarDigital Library
- R. Fagin and E. L. Wimmers. A Formula for Incorporating Weights into Scoring Rules. Theoretical Computer Science, 239(2):309--338, 2000. Google ScholarDigital Library
- A. Goldfarb and C. E. Tucker. Online Advertising, Behavioral Targeting, and Privacy. Commun. ACM, 54(5):25--27, 2011. Google ScholarDigital Library
- IMDB. IMDB Movie Ratings. http://www.imdb.com/interfaces.Google Scholar
- A. Machanavajjhala, E. Vee, M. Garofalakis, and J. Shanmugasundaram. Scalable Ranked Publish/Subscribe. In VLDB, 2008. Google ScholarDigital Library
- M. McGowan. Facebook Rolling Out Video Ads to News Feeds Social Network Gives Brands Four Demographics to Target with 15-Second Spots. Adweek. http://www.adweek.com/news/technology/facebook-rolling-out-video-ads-news-feeds-149239, May 2013.Google Scholar
- M. Prigg. Dislike: Over HALF of Facebook Users Say they are Fed Up with Constant Adverts and Sponsored Posts. Mail Online. (shortcut) http://unsourced.org/art/6978, July 2012.Google Scholar
- M. Sadoghi and H. Jacobsen. Relevance Matters: Capitalizing on Less (Top-k Matching in Publish/Subscribe). In ICDE, 2012. Google ScholarDigital Library
- C. Song, Z. Li, and T. Ge. Top-K Oracle: A New Way to Present Top-K Tuples for Uncertain Data. In ICDE, 2013.Google Scholar
- M. A. Stelzner. Social Media Marketing Industry Report. Social Media Examiner. http://www.socialmediaexaminer.com/SocialMediaMarketingReport2011.pdf, 2011.Google Scholar
- Yahoo! C15 - Yahoo! Music User Ratings of Musical Tracks, Albums, Artists and Genres v 1.0. Yahoo! Webscope http://webscope.sandbox.yahoo.com.Google Scholar
- J. Yan, N. Liu, G. Wang, W. Zhang, Y. Jiang, and Z. Chen. How Much Can Behavioral Targeting Help Online Advertising? In WWW, 2009. Google ScholarDigital Library
Recommendations
Fast Template Matching With Polynomials
Template matching is widely used for many applications in image and signal processing. This paper proposes a novel template matching algorithm, called algebraic template matching. Given a template and an input image, algebraic template matching ...
A fast bit-parallel multi-patterns string matching algorithm for biological sequences
ISB '10: Proceedings of the International Symposium on BiocomputingThe problem of searching occurrences of a pattern P[0...m-1] in the text T[0...n-1>with m ≤ n, where the symbols of P and T are drawn from some alphabet Σ of size σ, is called exact string matching problem. In the present day, pattern matching is a ...
A Dual-Bound Algorithm for Very Fast and Exact Template Matching
Recently proposed fast template matching techniques employ rejection schemes derived from lower bounds on the match measure. This paper generalizes that idea and shows that in addition to lower bounds, upper bounds on the match measure can be used to ...
Comments