Ridge regression for two dimensional locality preserving projection

Two Dimensional Locality Preserving Projection (2D-LPP) is a recent extension of LPP, a popular face recognition algorithm. It has been shown that 2D-LPP performs better than PCA, 2D-PCA and LPP. However, the computational cost of 2D-LPP is high. This paper proposes a novel algorithm called Ridge Regression for Two Dimensional Locality Preserving Projection (RR- 2DLPP), which is an extension of 2D-LPP with the use of ridge regression. RR-2DLPP is comparable to 2DLPP in performance whilst having a lower computational cost. The experimental results on three benchmark face data sets - the ORL, Yale and FERET databases - demonstrate the effectiveness and efficiency of RR-2DLPP compared with other face recognition algorithms such as PCA, LPP, SR, 2D-PCA and 2D-LPP.

Exploiting side information in locality preserving projection

Even if the class label information is unknown, side information represents some equivalence constraints between pairs of patterns, indicating whether pairs originate from the same class. Exploiting side information, we develop algorithms to preserve both the intra-class and inter-class local structures. This new type of locality preserving projection (LPP), called LPP with side information (LPPSI), preserves the data's local structure in the sense that the close, similar training patterns will be kept close, whilst the close but dissimilar ones are separated. Our algorithms balance these conflicting requirements, and we further improve this technique using kernel methods. Experiments conducted on popular face databases demonstrate that the proposed algorithm significantly outperforms LPP. Further, we show that the performance of our algorithm with partial side information (that is, using only small amount of pair-wise similarity/dissimilarity information during training) is comparable with that when using full side information. We conclude that exploiting side information by preserving both similar and dissimilar local structures of the data significantly improves performance.

Spike sorting using locality preserving projection with gap statistics and landmark-based spectral clustering

Understanding neural functions requires knowledge from analysing electrophysiological data. The process of assigning spikes of a multichannel signal into clusters, called spike sorting, is one of the important problems in such analysis. There have been various automated spike sorting techniques with both advantages and disadvantages regarding accuracy and computational costs. Therefore, developing spike sorting methods that are highly accurate and computationally inexpensive is always a challenge in the biomedical engineering practice.

Dynamic biometrics fusion at feature level for video-based human recognition

This paper proposes a novel human recognition method in video, which combines human face and gait traits
using a dynamic multi-modal biometrics fusion scheme. The Fisherface approach is adopted to extract face
features, while for gait features, Locality Preserving Projection (LPP) is used to achieve low-dimensional
manifold embedding of the temporal silhouette data derived from image sequences. Face and gait features are
fused dynamically at feature level based on a distance-driven fusion method. Encouraging experimental results
are achieved on the video sequences containing 20 people, which show that dynamically fused features produce
a more discriminating power than any individual biometric as well as integrated features built on common static
fusion schemes.

Neural signal analysis by landmark-based spectral clustering with estimated number of clusters

Spike sorting plays an important role in analysing electrophysiological data and understanding neural functions. Developing spike sorting methods that are highly accurate and computationally inexpensive is always a challenge in the biomedical engineering practice. This paper proposes an automatic unsupervised spike sorting method using the landmark-based spectral clustering (LSC) method in connection with features extracted by the locality preserving projection (LPP) technique. Gap statistics is employed to evaluate the number of clusters before the LSC can be performed. Experimental results show that LPP spike features are more discriminative than those of the popular wavelet transformation (WT). Accordingly, the proposed method LPP-LSC demonstrates a significant dominance compared to the existing method that is the combination between WT feature extraction and the superparamagnetic clustering. LPP and LSC are both linear algorithms that help reduce computational burden and thus their combination can be applied into realtime spike analysis.

Monotonicity preserving approximation of multivariate scattered data

This paper describes a new method of monotone interpolation and smoothing of multivariate scattered data. It is based on the assumption that the function to be approximated is Lipschitz continuous. The method provides the optimal approximation in the worst case scenario and tight error bounds. Smoothing of noisy data subject to monotonicity constraints is converted into a quadratic programming problem. Estimation of the unknown Lipschitz constant from the data by sample splitting and cross-validation is described. Extension of the method for locally Lipschitz functions is presented.

Guadalupian (Middle Permian) brachiopods from Sungai Toh, a Leptodus Shale locality in the Central Belt of Peninsular Malaysia. Part I: Lower Horizons

This work presents a systematic study of Permian Brachiopoda from the Sungai Toh Leptodus Shale locality, Pahang State, Peninsular Malaysia. This locality lies within the Central Belt of Peninsular Malaysia, a tectonic unit characterised.by tuffaceous sediments and limestones of Late Palaeozoic age. Two brachiopod-bearing horizons were studied in detail at this locality, the lower one (Horizon 2) bearing a mixed plant and invertebrate assemblage, including the brachiopods Urushtenoidea chaoi (CHING), Leptodus richthofeni KAYSER, Anidanthus cf. sinosus HUANG, Acosarina dorashamensis (SOKOLSKAJA), A. minuta (ABleH) and unidentifiable species of Linoproduetus, Neochonetes, and Strophalosiina. Horizon 3 contains a more abundant and diverse brachiopod fauna, comprising a total. of 57 species representing 47 genera, including Vediproductus punetatiformis (CHAO), Permianella typica HE & ZHU, Tranrennatia gratiosa (WAAGEN), Leptodus richthofeni KAYSER, Leptodus cf. tenuis (WAAGEN) and "Semibrachythyrina" [= Alphaneospirifer] cf. pyramidiformis LIANG. It is
suggested in this study that the age of the Sungai Toh locality is Capitanian (late Guadalupian) to possibly Wuchiapingian (early Lopingian),
as it appears to correlate well with the Lengwu fauna from Zhejiang in eastern China. The palaeobiogeographical affinities of the Sungai Toh fauna are interesting, mainly indicating strong Palaeo-equatorial affinities, while there are also some elements more typical of the cooler periGondwana
Region.

Projections of the future path of the gender wage gap in Great Britain

Purpose – The paper attempts to project the future trend of the gender wage gap in Great Britain up to 2031.
Design/methodology/approach – The empirical analysis utilises the British Household Panel Study Wave F together with Office for National Statistics (ONS) demographic projections. The methodology combines the ONS projections with assumptions relating to the evolution of educational attainment in order to project the future distribution of human capital skills and consequently the future size of the gender wage gap.
Findings – The analysis suggests that gender wage convergence will be slow, with little female progress by 2031 unless there is a large rise in returns to female experience.
Originality/value – The paper has projected the pattern of male and female skill acquisition together with the associated trend in wages up to 2031.

Nursing education and graduates : profiles for 1999 and 2000, with projections for 2001

An analysis of DEST statistical data on nursing education had suggested that there were issues with the classification of undergraduate data and new initiatives, such as combined degrees, which were poorly captured. This project had two distinct aims. First, to explore and map exhaustively the range of undergraduate programs offered by tertiary education providers across Australia leading to an initial qualification and entry into nursing practice. Second, to explore and map in detail specialist nursing education courses offered by tertiary and other education providers across Australia. This research was funded under the Evaluations and Investigations Programme (EIP).

Nursing educations and graduates : profiles for 2001 and projections for 2002

The present study was commissioned for the National Review of Nursing Education. This is the second of two national studies commissioned to map in detail nursing education programs and to profile and make future projections regarding graduates from undergraduate and postgraduate nursing education courses in Australia.

The first study was undertaken in 2001 by Deakin University School of Nursing under the auspices of the 2001 Evaluation Investigations Project titled "Nursing Education and Graduates: Profiles for 1999, and 2000 with projections for 2001". This project sought data on nursing education within Australia in order to improve the accuracy of nursing education databases and to strengthen the ability of DETYA to provide advice on workforce planning. Issues that arose from that project included differences in data sets for undergraduate nursing courses in Australia and the complex process of attempting to tease out and accurately quantify postgraduate specialty courses when a trend towards postgraduate generic courses was evident. Approximately 26% of postgraduate domestic student enrolment data were reported utilising a generic nursing course category.

The purpose of this study was, therefore, twofold. Firstly, this study validated and extended the existing database developed in the previous study mapping in detail the full range of undergraduate programs offered by tertiary education providers across Australia that lead to an initial qualification and entry into nursing practice.

New data about the following was sought:

* Undergraduate nursing degrees (both three-year and four-year courses);
* Double/combined nursing degrees;
* Courses offered by private universities;
* Four-year bachelor degrees that concurrently provide both initial nurse registration and preparation for specialty nursing practice;
* Courses that facilitate ‘fast-tracking’ of students for initial nurse registration with previous tertiary or nursing studies,
* Hours and configuration of clinical experience in undergraduate nursing courses.

Nursing education and graduates in Victoria : profiles for 2000, 2001, and 2002 with projections for 2003

This project is the second of two projects commissioned by the Nurse Policy Branch, Victorian Department of Human Services (VDHS). The project investigated nursing education at the undergraduate and postgraduate levels in the university sector and in the VET sector. The data gathered from the project provided VDHS with a better understanding of nurse labour force issues.

Shape preserving approximation using least squares splines

Least squares polynomial splines are an effective tool for data fitting, but they may fail to preserve essential properties of the underlying function, such as monotonicity or convexity. The shape restrictions are translated into linear inequality conditions on spline coefficients. The basis functions are selected in such a way that these conditions take a simple form, and the problem becomes non-negative least squares problem, for which effecitive and robust methods of solution exist. Multidimensional monotone approximation is achieved by using tensor-product splines with the appropriate restrictions. Additional inter polation conditions can also be introduced. The conversion formulas to traditional B-spline representation are provided.