Concurrent multiresponse non-linear screening: Robust profiling of webpage performance

Profiling engineered data with robust mining methods continues attracting attention in knowledge engineering systems. The purpose of this article is to propose a simple technique that deals with non-linear multi-factorial multi-characteristic screening suitable for knowledge discovery studies. The method is designed to proactively seek and quantify significant information content in engineered mini-datasets. This is achieved by deploying replicated fractional-factorial sampling schemes. Compiled multi-response data are converted to a single master-response effectuated by a series of distribution-free transformations and multi-compressed data fusions. The resulting amalgamated master response is deciphered by non-linear multi-factorial stealth stochastics intended for saturated schemes. The stealth properties of our method target processing datasets which might be overwhelmed by a lack of knowledge about the nature of reference distributions at play. Stealth features are triggered to overcome restrictions regarding the data normality conformance, the effect sparsity assumption and the inherent collapse of the ‘unexplainable error’ connotation in saturated arrays. The technique is showcased by profiling four ordinary controlling factors that influence webpage content performance by collecting data from a commercial browser monitoring service on a large scale web host. The examined effects are: (1) the number of Cascading Style Sheets files, (2) the number of JavaScript files, (3) the number of Image files, and (4) the Domain Name System Aliasing. The webpage performance level was screened against three popular characteristics: (1) the time to first visual, (2) the total loading time, and (3) the customer satisfaction. Our robust multi-response data mining technique is elucidated for a ten-replicate run study dictated by an L₉(3⁴) orthogonal array scheme where any uncontrolled noise embedded contribution has not been necessarily excluded.     

SIMS as a subnanometer probe: A new tool for chemical profile analysis of grafted molecules

The complexity of modern engineered surfaces requires the development of very powerful methods to analyze and characterize them. We demonstrate that it is possible to obtain chemical information about the skeleton of organic molecules constituting SAMs grafted on a silicon surface by using a new type of SIMS method. A profile can be achieved by the investigation of the temporal variation of secondary ion intensities that correspond to the fractional parts of the molecule constituting the SAMs. The equivalent ablation rate is less than 0.5 nm/min.     

Quantitative Depth Profiling of K-Doped Fullerene Films Using XPS and SIMS

 Phenomena accompanying electrochemical doping of solid fullerene films with potassium were studied by sputter ion depth profiling (XPS and SIMS). The potassium distribution was determined, and artifacts associated with possible damage of the layer composition caused by ion impact were investigated and discussed. To compare the charge transfer while reductive doping is taking place at fullerene/solution interface with doping from gas phase, model layers were prepared and doped by potassium under UHV conditions. It was found that sputtering by Ar⁺ primary ions yields both accurate information on the alkaline metal distribution and on its concentration. Sputtering by O⁺ ions led to an enrichment of potassium, apparently due to the reactivity of oxygen with the fullerene matrix. It is shown that the reductive doping starts at the fullerene/solution interface. The concentration of potassium in the doped films was found to be lower than expected from the charge transferred during the electrochemical reduction. Other phase transformations such as hydrogenation are discussed. Received March 4, 2002; accepted July 26, 2002     

Constructing Space for Science at the Royal Institution of Great Britain

Those who have worked in the Royal Institution of Great Britain have, since its foundation in 1799, made significant contributions to scientific knowledge, to its practical application, and to its communication to a wide variety of audiences. Such work cannot be carried out in an architectural vacuum, and in this paper we examine how the buildings of the Royal Institution, 20 and 21 Albemarle Street in central London, have shaped the work undertaken within its walls and how, on a number of occasions, the buildings have been reconfigured to take account of the evolving needs of scientific research and communication. This paper is based on the Conservation Plan of the Royal Institution that we wrote during 2003. The Conservation Plan did not examine the land owned by the Royal Institution to the north (i.e., 22 and 23 Albemarle Street; for this area see Richard Garnier, “Grafton Street, Mayfair,” Georgian Group Journal 13 (2003), 210–272), but it did discuss 18 and 19 Albemarle Street. In this paper we concentrate on the core Royal Institution buildings at 20 and 21 Albemarle Street. Other studies of the relationship of architecture,space, and science include Crosbie Smith and Jon Agar, ed., Making Space for Science: Territorial Themes in the Shaping of Knowledge (Basingstoke: Macmillan, 1997); Peter Galison and Emily Thompson, ed., The Architecture of Science (Cambridge, Mass.: MIT Press, 1999); and Sophie Forgan,“The architecture of science and the idea of a university,” Studies in History and Philosophy of Science 20 (1989), 405–434. Frank A.J.L. James is Professor of the History of Science at the Royal Institution; he has written widely on the history of nineteenth-century science in its social and cultural contexts and is editor of the Correspondence of Michael Faraday. He is President of the British Society for the History of Science. Anthony Peers is an Associate of Rodney Melville and Partners where he works in the field of building conservation as an architectural historian. He is a Council member of the Ancient Monument Society.     

Extension of multivariate regression trees to interval data. Application to electricity load profiling

Summary  Several data can be presented as interval curves where intervals reflect a within variability. In particular, this representation is well adapted for load profiles, which depict the electricity consumption of a class of customers. Electricity load profiling consists in assigning a daily load curve to a customer based on their characteristics such as energy requirement. Within the load profiling scope, this paper investigates the extension of multivariate regression trees to the case of interval dependent (or response) variables. The tree method aims at setting up simultaneously load profiles and their assignment rules based on independent variables. The extension of multivariate regression trees to interval responses is detailed and a global approach is defined. It consists in a first stage of a dimension reduction of the interval response variables. Thereafter, the extension of the tree method is applied to the first principal interval components. Outputs are the classes of the interval curves where each class is characterized both by an interval load profile (e.g. the class prototype) and an assignment rule based on the independent variables.