Search results for "dimensionality"

Applying fully tensorial ICA to fMRI data

2016

There are two aspects in functional magnetic resonance imaging (fMRI) data that make them awkward to analyse with traditional multivariate methods - high order and high dimension. The first of these refers to the tensorial nature of observations as array-valued elements instead of vectors. Although this can be circumvented by vectorizing the array, doing so simultaneously loses all the structural information in the original observations. The second aspect refers to the high dimensionality along each dimension making the concept of dimension reduction a valuable tool in the processing of fMRI data. Different methods of tensor dimension reduction are currently gaining popUlarity in literature…

Using affinity perturbations to detect web traffic anomalies

2013

The initial training phase of machine learning algorithms is usually computationally expensive as it involves the processing of huge matrices. Evolving datasets are challenging from this point of view because changing behavior requires updating the training. We propose a method for updating the training profile efficiently and a sliding window algorithm for online processing of the data in smaller fractions. This assumes the data is modeled by a kernel method that includes spectral decomposition. We demonstrate the algorithm with a web server request log where an actual intrusion attack is known to happen. Updating the kernel dynamically using a sliding window technique, prevents the proble…

Beyond Tandem Analysis: Joint Dimension Reduction and Clustering in R

2019

We present the R package clustrd which implements a class of methods that combine dimension reduction and clustering of continuous or categorical data. In particular, for continuous data, the package contains implementations of factorial K-means and reduced K-means; both methods combine principal component analysis with K-means clustering. For categorical data, the package provides MCA K-means, i-FCB and cluster correspondence analysis, which combine multiple correspondence analysis with K-means. Two examples on real data sets are provided to illustrate the usage of the main functions.

Factorial Structure and Psychometric Properties of the Sensation Seeking Scale – Form V (SSS-V) in a Sample of Italian Adolescents

2013

The present study was designed to evaluate the psychometric properties of the Sensation Seeking Scale – Form V (SSS-V) for the Italian-speaking population. The instrument was administered to 1,530 high school and college students. A second group of 268 high school students completed the SSS-V and the Impulsiveness Questionnaire (IVE). Exploratory factor analysis highlighted a 2-factor structure, Thrill and Adventure Seeking (TAS) and Disinhibition (Dis). Confirmatory factor analysis revealed an adequate model fit. Internal consistency of the subscales was marginally supported using the alpha reliability measure. Convergent validity was supported by significant positive correlations between …

The Importance of Electronic Dimensionality in Multiorbital Radical Conductors

2019

The exceptional performance of oxobenzene-bridged bis-1,2,3-dithiazolyls 6 as single-component neutral radical conductors arises from the presence of a low-lying π-lowest unoccupied molecular orbital, which reduces the potential barrier to charge transport and increases the kinetic stabilization energy of the metallic state. As part of ongoing efforts to modify the solid-state structures and transport properties of these so-called multiorbital materials, we report the preparation and characterization of the acetoxy, methoxy, and thiomethyl derivatives 6 (R = OAc, OMe, SMe). The crystal structures are based on ribbonlike arrays of radicals laced together by S···N′ and S···O′ secondary bondin…

Application of Electronic Nose for Evaluation of Wastewater Treatment Process Effects at Full-Scale WWTP

2019

This paper presents the results of studies aiming at the assessment and classification of wastewater using an electronic nose. During the experiment, an attempt was made to classify the medium based on an analysis of signals from a gas sensor array, the intensity of which depended on the levels of volatile compounds in the headspace gas mixture above the wastewater table. The research involved samples collected from the mechanical and biological treatment devices of a full-scale wastewater treatment plant (WWTP), as well as wastewater analysis. The measurements were carried out with a metal-oxide-semiconductor (MOS) gas sensor array, when coupled with a computing unit (e.g., a computer with…

Communication between iron(II) building blocks in cooperative spin transition phenomena

2003

[EN] In the present article we discuss the cooperative nature of the spin crossover phenomenon in iron(II) complexes, providing a perspective of the state of the art in this area. The first aspect we discuss is the role of the intermolecular interactions, more precisely the ¿-interactions, in mononuclear complexes. We show that by playing with the nature of the ligands, aliphatic, aromatic, or extended aromatic, it is possible to create stronger cohesive forces and receive a more cooperative response from the compound. In the next step the singular family of bipyrimidine-bridged iron(II) dinuclear compounds is presented as the simplest example of polynuclear spin crossover complexes exhibit…

Some Considerations on 3-D and 2-D Numerical Models for the Assessment of the Stability of Underground Caves

2014

The application of numerical modeling to the analysis of the stability of both natural and man-made underground caves is rapidly increasing due to the availability of powerful numerical codes, that can account for either continuum or discontinuum behavior of the rock masses. Numerical methods allow to overcome traditional methods for cave stability analysis that assume too simplified geometrical, geological and geomechanical conditions. Further, they are also able to assess the potential failure mechanisms of underground systems. On the other hand, the application of numerical methods requires availability of a detailed geo-structural survey of the cave, as well as a proper geomechanical ch…

Large-scale nonlinear dimensionality reduction for network intrusion detection

2017

International audience; Network intrusion detection (NID) is a complex classification problem. In this paper, we combine classification with recent and scalable nonlinear dimensionality reduction (NLDR) methods. Classification and DR are not necessarily adversarial, provided adequate cluster magnification occurring in NLDR methods like $t$-SNE: DR mitigates the curse of dimensionality, while cluster magnification can maintain class separability. We demonstrate experimentally the effectiveness of the approach by analyzing and comparing results on the big KDD99 dataset, using both NLDR quality assessment and classification rate for SVMs and random forests. Since data involves features of mixe…

Knowledge discovery using diffusion maps

2013