Search results for "Analysis"

Online anomaly detection using dimensionality reduction techniques for HTTP log analysis

2015

Modern web services face an increasing number of new threats. Logs are collected from almost all web servers, and for this reason analyzing them is beneficial when trying to prevent intrusions. Intrusive behavior often differs from the normal web traffic. This paper proposes a framework to find abnormal behavior from these logs. We compare random projection, principal component analysis and diffusion map for anomaly detection. In addition, the framework has online capabilities. The first two methods have intuitive extensions while diffusion map uses the Nyström extension. This fast out-of-sample extension enables real-time analysis of web server traffic. The framework is demonstrated using …

Justification of point electrode models in electrical impedance tomography

2011

The most accurate model for real-life electrical impedance tomography is the complete electrode model, which takes into account electrode shapes and (usually unknown) contact impedances at electrode-object interfaces. When the electrodes are small, however, it is tempting to formally replace them by point sources. This simplifies the model considerably and completely eliminates the effect of contact impedance. In this work we rigorously justify such a point electrode model for the important case of having difference measurements ("relative data") as data for the reconstruction problem. We do this by deriving the asymptotic limit of the complete model for vanishing electrode size. This is s…

SaaS architecture and pricing models

2014

In the new era of computing, SaaS software with different architectural characteristics might be priced in different ways. Even though both pricing and architectural characteristics are responsible for the success of the offering; the relationship between architectural and pricing characteristics has not been studied before. The present study fills this gap by employing a multi-case research. The findings accentuate that flexible and well-designed architecture enables different pricing models; however, poorly designed architecture limits also the pricing. Scalability and high level of modularity are the major enablers of a great variety of pricing models. Using public cloud services may lea…

Efficient Time Integration of Maxwell's Equations with Generalized Finite Differences

2015

We consider the computationally efficient time integration of Maxwell’s equations using discrete exterior calculus (DEC) as the computational framework. With the theory of DEC, we associate the degrees of freedom of the electric and magnetic fields with primal and dual mesh structures, respectively. We concentrate on mesh constructions that imitate the geometry of the close packing in crystal lattices that is typical of elemental metals and intermetallic compounds. This class of computational grids has not been used previously in electromagnetics. For the simulation of wave propagation driven by time-harmonic source terms, we provide an optimized Hodge operator and a novel time discretizati…

Detection of Fake Profiles in Social Media - Literature Review

2017

Research literature clustering using diffusion maps

2013

We apply the knowledge discovery process to the mapping of current topics in a particular field of science. We are interested in how articles form clusters and what are the contents of the found clusters. A framework involving web scraping, keyword extraction, dimensionality reduction and clustering using the diffusion map algorithm is presented. We use publicly available information about articles in high-impact journals. The method should be of use to practitioners or scientists who want to overview recent research in a field of science. As a case study, we map the topics in data mining literature in the year 2011. peerReviewed

Revealing Fake Profiles in Social Networks by Longitudinal Data Analysis

2017

Short-Time Non-work-related Computing and Creative Performance

2014

It has been argued that non-work-related computing (NWRC) takes time away from work and, hence, decreases work productivity. On the other hand, it has also been claimed that short-time non-work-related computing (STNWRC) (a maximum of 15 minutes), has a positive impact on work productivity, including relief from boredom, higher creativity, and the underlying recovery mechanisms. To examine the impact of STNWRC on creative performance, we draw on Fredrickson's broaden-and-build theory, the concept of recovery with mental well-being and low cognitive effort. A 2 × 2 factorial experiment with 40 subjects was conducted. The results indicate that STNWRC has a positive effect on creative performa…

Affective contents of cross-cultural audiovisual experience

2015

Audiovisual presentation of a product has a direct impact on the mental representation of an individual when interacting with a product. Companies produce audiovisual contents that can be used in different cultural environments as a way of having broader impact at a cheaper cost. But do video contents have the same impact in different countries? We addressed this question by using images and videos of Iittala products (Finnish design) with the goal of finding whether participants from two countries, Finland and Spain, appraised the designs similarly. We performed an experiment in which participants interacted with three audiovisual products depicting images/videos of the brand design. Throu…

Determining the number of sources in high-density EEG recordings of event-related potentials by model order selection

2011

To high-density electroencephalography (EEG) recordings, determining the number of sources to separate the signal and the noise subspace is very important. A mostly used criterion is that percentage of variance of raw data explained by the selected principal components composing the signal space should be over 90%. Recently, a model order selection method named as GAP has been proposed. We investigated the two methods by performing independent component analysis (ICA) on the estimated signal subspace, assuming the number of selected principal components composing the signal subspace is equal to the number of sources of brain activities. Through examining wavelet-filtered EEG recordings (128…