Search results for "Data mining"
showing 10 items of 907 documents
Vector Autoregressive Fractionally Integrated Models to Assess Multiscale Complexity in Cardiovascular and Respiratory Time Series
2020
Cardiovascular variability is the result of the activity of several physiological control mechanisms, which involve different variables and operate across multiple time scales encompassing short term dynamics and long range correlations. This study presents a new approach to assess the multiscale complexity of multivariate time series, based on linear parametric models incorporating autoregressive coefficients and fractional integration. The approach extends to the multivariate case recent works introducing a linear parametric representation of multiscale entropy, and is exploited to assess the complexity of cardiovascular and respiratory time series in healthy subjects studied during postu…
Exploiting Deductive Processes for Automated Network Management
2005
This paper focuses on improving network management by the adoption of artificial intelligence techniques. We propose a distributed multiagent architecture for network management, which exploits the dynamic reasoning capabilities of the situation calculus in order to emulate the reactive behavior of a human expert to fault situations. The information related to network events is generated by programmable sensors deployed on the network devices and is collected by a logical entity for network managing where it is merged with general domain knowledge, with a view to identifying the root causes of faults and to decide on reparative actions. The logical inference system has been devised to carry…
Learning Temporal Regularities of User Behavior for Anomaly Detection
2001
Fast expansion of inexpensive computers and computer networks has dramatically increased number of computer security incidents during last years. While quite many computer systems are still vulnerable to numerous attacks, intrusion detection has become vitally important as a response to constantly increasing number of threats. In this paper we discuss an approach to discover temporal and sequential regularities in user behavior. We present an algorithm that allows creating and maintaining user profiles relying not only on sequential information but taking into account temporal features, such as events' lengths and possible temporal relations between them. The constructed profiles represent …
Clustering Bacteria Species Using Neural Gas: Preliminary Study
2009
In this work a method for clustering and visualization of bacteria taxonomy is presented. A modified version of the Batch Median Neural Gas (BNG) algorithm is proposed. The BNG algorithm is able to manage non vectorial data given as a dissimilarity matrix. We tested the modified BNG on the dissimilarity matrix obtained from sequences alignment and computing distances using bacteria genotype information regarding the16S rRNA housekeeping gene, which represents a stable part of bacteria genome. The dataset used for the experiments is obtained from the Ribosomal Database Project II, and it is made of 5159 sequences of 16S rRNA genes. Preliminary results of the experiments show a promising abil…
A Comparative Analysis of Residual Block Alternatives for End-to-End Audio Classification
2020
Residual learning is known for being a learning framework that facilitates the training of very deep neural networks. Residual blocks or units are made up of a set of stacked layers, where the inputs are added back to their outputs with the aim of creating identity mappings. In practice, such identity mappings are accomplished by means of the so-called skip or shortcut connections. However, multiple implementation alternatives arise with respect to where such skip connections are applied within the set of stacked layers making up a residual block. While residual networks for image classification using convolutional neural networks (CNNs) have been widely discussed in the literature, their a…
In-depth evaluation of software tools for data-independent acquisition based label-free quantification.
2015
Label-free quantification (LFQ) based on data-independent acquisition workflows currently experiences increasing popularity. Several software tools have been recently published or are commercially available. The present study focuses on the evaluation of three different software packages (Progenesis, synapter, and ISOQuant) supporting ion mobility enhanced data-independent acquisition data. In order to benchmark the LFQ performance of the different tools, we generated two hybrid proteome samples of defined quantitative composition containing tryptically digested proteomes of three different species (mouse, yeast, Escherichia coli). This model dataset simulates complex biological samples con…
BANΔIT: B’‐factor Analysis for Drug Design and Structural Biology
2020
The analysis of B‐factor profiles from X‐ray protein structures can be utilized for structure‐based drug design since protein mobility changes have been associated with the quality of protein‐ligand interactions. With the BANΔIT (B’‐factor analysis and ΔB’ interpretation toolkit), we have developed a JavaScript‐based browser application that provides a graphical user interface for the normalization and analysis of B’‐factor profiles. To emphasize the usability for rational drug design applications, we have analyzed a selection of crystallographic protein‐ligand complexes and have given exemplary conclusions for further drug optimization including the development of a B’‐factor‐supported pha…
On a new robust workflow for the statistical and spatial analysis of fracture data collected with scanlines (or the importance of stationarity)
2020
Abstract. We present an innovative workflow for the statistical analysis of fracture data collected along scanlines, composed of two major stages, each one with alternative options. A prerequisite in our analysis is the assessment of stationarity of the dataset, which is motivated by statistical and geological considerations. Calculating statistics on non-stationary data can be statistically meaningless, and moreover the normalization and/or sub-setting approach that we discuss here can greatly improve our understanding of geological deformation processes. Our methodology is based on performing non-parametric statistical tests, which allow detecting important features of the spatial distrib…
A quantitative analysis of Educational Data through the Comparison between Hierarchical and Not-Hierarchical Clustering
2017
Many research papers have studied the problem of taking a set of data and separating it into subgroups through the methods of Cluster Analysis. However, the variables and parameters involved in Cluster Analysis have not always been outlined and criticized, especially in the field of Science Education. Moreover, in the field of Science Education, a comparison between two different Clustering methods is not discussed in the literature. Conceptions of students about modeling in physic are investigated by using an open-ended questionnaire. The questionnaire is analyzed through Clustering methods. The clustering results obtained by using the two methods are compared and show a good coherence bet…
A Novel Technique for Fingerprint Classification based on Fuzzy C-Means and Naive Bayes Classifier
2014
Fingerprint classification is a key issue in automatic fingerprint identification systems. One of the main goals is to reduce the item search time within the fingerprint database without affecting the accuracy rate. In this paper, a novel technique, based on topological information, for efficient fingerprint classification is described. The proposed system is composed of two independent modules: the former module, based on Fuzzy C-Means, extracts the best set of training images, the latter module, based on Fuzzy C-Means and Naive Bayes classifier, assigns a class to each processed fingerprint using only directional image information. The proposed approach does not require any image enhancem…