Faculty of Informatics - Papers (Archive)

Generalised score and Wald tests

Paul Rippon — Sun, 14 Apr 2013 15:39:50 PDT

The generalised score and Wald tests are described and related to their nongeneralised versions. Two interesting applications are discussed. In the first a new test for the Behrens-Fisher problem is derived. The second is testing homogeneity of variances from multiple univariate normal populations.

Connecting museum collections and creator communities: The Virtual Museum of the Pacific project

Frank Howarth — Wed, 03 Apr 2013 17:05:04 PDT

[extract] “I’m interested in what you can’t see in collections...where this has been, what its function could be and how it still exists today...artists have a perspective that may assist in keeping some of these things in the way these objects always were made, but [also] ensuring that they evolve as well because that's part of the culture”. Artist Latai Taumopeau talking about a Ngatu, a traditional bark cloth from Tonga, in the Australian Museum collection.

A five-round algebraic property of AES and its application to the ALPHA-MAC

Jianyong Huang — Mon, 18 Mar 2013 17:12:44 PDT

We present a five-round algebraic property of the advanced encryption standard (AES), and we show that this algebraic property can be used to analyse the internal structure of ALPHA-MAC whose underlying block cipher is AES. In the proposed property, we modify 20 bytes from five intermediate values at some fixed locations in five consecutive rounds, and we show that after five rounds of operations, such modifications do not change the intermediate result and finally, still produce the same ciphertext. By employing the proposed five-round algebraic property of AES, we provide a method to find second preimages of the ALPHA-MAC based on the assumption that a key or an intermediate value is known. We also show that our idea can also be used to find internal collisions of the ALPHA-MAC under the same assumption.

An ontology based approach for health information discovery

Khin T. Win — Sun, 03 Mar 2013 17:07:58 PST

Short designated verifier signature scheme and its identity-based variant

Xinyi Huang — Sun, 24 Feb 2013 21:51:19 PST

The notion of strong designated verifier signature was put forth by Jakobsson, Sako and Impagliazzo in 1996, but the formal definition was defined recently by Saeednia, Kremer and Markowitch in 2003 and revisited by Laguil- laumie and Vergnaud in 2004. In this paper, we firstly propose the notion of short strong designated verifier sig- nature scheme, and extend it to the short identity-based strong designated verifier scheme. Then, we propose the first construction of short strong designated verifier sig- nature scheme. We also extend our scheme to construct a short identity-based strong designated verifier signature scheme. The size of the signature of our schemes is the shortest compared to any existing schemes reported in the literature. We provide formal security proofs for our schemes based on the random oracle model. Finally, we also discuss an extension of our scheme to construct a short strong designated verifier signature without random oracle.

Kriging nonstationary data

Noel A. Cressie — Tue, 19 Feb 2013 22:29:34 PST

Spatial data modeled to have come from a random function with a nonstationary mean are considered. The spatial prediction method known as kriging exploits second-order spatial correlation structure to obtain minimum variance predictions of certain average values of the random function. But to do so, it must be assumed that either the mean function (the drift) is known up to a constant or the second-order structure (the variogram) is known exactly. Knowledge of the drift allows the (stationary) variogram to be estimated and leads to ordinary kriging. Knowledge of the variogram allows the drift to be estimated and leads to universal kriging. More usually, neither is known. This article shows how median polish of gridded spatial data provides a resistant and relatively bias-free way of kriging in the presence of drift, yet yields results as good as the mathematically optimal (but operationally difficult) universal kriging. Comparisons are performed on two data sets.

A Useful Empirical Bayes Identity

Noel A. Cressie — Tue, 19 Feb 2013 20:50:23 PST

Playing safe with misweighted means

Noel A. Cressie — Tue, 19 Feb 2013 20:45:23 PST

The moment-generating function and negative integer moments

Noel A. Cressie — Tue, 19 Feb 2013 20:45:23 PST

The asymptotic-distribution of the scan statistic under uniformity

Noel A. Cressie — Tue, 19 Feb 2013 20:40:23 PST

Empirical bayes estimation for discrete-distributions

Noel A. Cressie — Tue, 19 Feb 2013 20:40:22 PST

The supremum distribution of another gaussian process

Noel A. Cressie — Tue, 19 Feb 2013 20:30:35 PST

Transformations and the jackknife

Noel A. Cressie — Tue, 19 Feb 2013 20:30:35 PST

The woolf plot is more reliable than the scatchard plot in analyzing data from hormone receptor assays

D Keightley — Tue, 19 Feb 2013 20:30:35 PST

Random set limit-theorems

Noel A. Cressie — Tue, 19 Feb 2013 20:30:33 PST

Robust estimation of the variogram 1

Noel A. Cressie — Tue, 19 Feb 2013 20:25:24 PST

Exponential and power data transformations

Noel A. Cressie — Tue, 19 Feb 2013 20:25:23 PST

Strong limit-theorem for random sets

Noel A. Cressie — Tue, 19 Feb 2013 20:25:23 PST

Some properties of scan statistic on circle and line

Noel A. Cressie — Tue, 19 Feb 2013 20:20:24 PST

Empirical bayes procedure for finding an interval estimate

F Lord — Tue, 19 Feb 2013 20:20:24 PST