I
Ioana A. Cosma
Researcher at University of Ottawa
Publications - 8
Citations - 210
Ioana A. Cosma is an academic researcher from University of Ottawa. The author has contributed to research in topics: Estimator & Probabilistic analysis of algorithms. The author has an hindex of 5, co-authored 8 publications receiving 198 citations. Previous affiliations of Ioana A. Cosma include University of Oxford & University of Cambridge.
Papers
More filters
Journal ArticleDOI
Estimating Sizes of Social Networks via Biased Sampling
TL;DR: This article presents algorithms for estimating the number of users in online social networks by adopting the standard abstraction of social networks as undirected graphs and performing random walk-based node sampling, showing analytically that the estimate error vanishes with high probability for fewer samples than those required by prior-art algorithms.
Journal ArticleDOI
A Statistical Analysis of Probabilistic Counting Algorithms
Peter Clifford,Ioana A. Cosma +1 more
TL;DR: This article applies conventional statistical methods to compare probabilistic algorithms based on storing either selected order statistics, or random projections, and derives estimators of the cardinality in both cases, and shows that the maximal‐term estimator is recursively computable and has exponentially decreasing error bounds.
Proceedings Article
A simple sketching algorithm for entropy estimation over streaming data.
Peter Clifford,Ioana A. Cosma +1 more
TL;DR: It is shown that the random variables used in estimating the Renyi entropy can be transformed to have a proper distributional limit as α approaches 1, and a family of asymptotically unbiased log-mean estimators of the Shannon entropy, indexed by a constant ζ > 0, are proposed.
Posted Content
A statistical analysis of probabilistic counting algorithms
Peter Clifford,Ioana A. Cosma +1 more
TL;DR: In this paper, a statistical analysis of probabilistic counting algorithms for cardinality estimation in data stream applications is presented, focusing on two techniques that use pseudo-random variates to form low-dimensional data sketches.
Dissertation
Dimension reduction of streaming data via random projections
TL;DR: It is argued that online, fast, and efficient computation of summary statistics such as cardinality, entropy, and l_{alpha} distances may be a useful qualitative tool for detecting lack of convergence, and illustrated with simulations of the posterior distribution of a decomposable Gaussian graphical model via the Metropolis-Hastings algorithm.