Classification of text documents presents a unique challenge to conventional classification algorithms. Due to the existence of large number of features in the datasets, providing a desired representation for text documents can be seen as another problem. In this paper a simple but effective representation model for text documents to tackle the classification problem is discussed. Two different...

This paper details the results of a face verification competition [2] held in conjunction with the Second International Conference on Biometric Authentication. The contest was held on the publically available XM2VTS database [4] according to a defined protocol [15]. The aim of the competition was to assess the advances made in face recognition since 2003 and to measure the sensitivity of the te...

We introduce a computer vision problem from social cognition, namely, the automated detection of attitudes from a person's spontaneous facial expressions. To illustrate the challenges, we introduce two simple algorithms designed to predict observers' preferences between images (e.g., of celebrities) based on covert videos of the observers' faces. The two algorithms are almost as accurate as hum...

Behavioral displays or physiological responses are often influenced by intrinsic and extrinsic mechanisms in the context of the organism's evolutionary history. Understanding differences in transcriptome profiles can give insight into adaptive or pathological responses. We utilize high throughput sequencing (RNA-sequencing) to characterize the neurotranscriptome profiles in both males and femal...

This paper provides an introduction to specifying multilevel models using PROC MIXED. After a brief introduction to the field of multilevel modeling, users are provided with concrete examples of how PROC MIXED can be used to estimate (a) two-level organizational models, (b) two-level growth models, and (c) three-level organizational models. Both random intercept and random intercept and slope m...

In this paper, we compare and evaluate five contemporary, data-driven real-time 2D object tracking methods: the region tracker by Hager et al., the Hyperplane tracker, the CONDENSATION tracker, and the Mean Shift and Trust Region trackers. The first two are classical template based methods, while the latter three are from the more recently proposed class of histogram based trackers. All tracker...

We point out the existence of a disturbing deficiency in the field of system identification, namely the fact that many results, published in papers, are not reproducible. In many cases, datasets and time series, that are used to illustrate identification methods and algorithms in these publications, are not freely available. We propose to remedy this serious deficiency by setting up a publicall...

We investigate the combined use of multiple structuring elements with the standard morphological texture characterization tools, namely morphological covariance and granulometry. The resulting operator is applied to both grayscale and color images in the context of texture classi cation. As to its extension to color texture data, it is realized by means of a weighting based reduced vector order...

As internet users query search engines, relevant advertisements accompany the search results. Advertisers bid for limited ad spots in a generalized-secondary auction employed by search engines. This paper uses publically-available data from Google to test strategies advertisers may use when deciding where and how much to bid for a keyword phrase. A predictive model of suggested bid price may pr...

DO-178B was based on the consensus of the avionic software community as it existed in 1992. Twenty two years after publication, we have no publically available experimental data as to its efficacy. It appears to work extremely well, since there have been no hull loss accidents in passenger service ascribed to software failure. This is a comforting and surprising result. However, if we don’t kno...