Search results for "DOM"
showing 10 items of 12668 documents
Application of selected supervised classification methods to bank marketing campaign
2016
Supervised classification covers a number of data mining methods based on training data. These methods have been successfully applied to solve multi-criteria complex classification problems in many domains, including economical issues. In this paper we discuss features of some supervised classification methods based on decision trees and apply them to the direct marketing campaigns data of a Portuguese banking institution. We discuss and compare the following classification methods: decision trees, bagging, boosting, and random forests. A classification problem in our approach is defined in a scenario where a bank’s clients make decisions about the activation of their deposits. The obtained…
Biological and Mechanical Characterization of the Random Positioning Machine (RPM) for Microgravity Simulations
2021
The rapid improvement of space technologies is leading to the continuous increase of space missions that will soon bring humans back to the Moon and, in the coming future, toward longer interplanetary missions such as the one to Mars. The idea of living in space is charming and fascinating; however, the space environment is a harsh place to host human life and exposes the crew to many physical challenges. The absence of gravity experienced in space affects many aspects of human biology and can be reproduced in vitro with the help of microgravity simulators. Simulated microgravity (s-μg) is applied in many fields of research, ranging from cell biology to physics, including cancer biology. In…
Improving Scalable K-Means++
2021
Two new initialization methods for K-means clustering are proposed. Both proposals are based on applying a divide-and-conquer approach for the K-means‖ type of an initialization strategy. The second proposal also uses multiple lower-dimensional subspaces produced by the random projection method for the initialization. The proposed methods are scalable and can be run in parallel, which make them suitable for initializing large-scale problems. In the experiments, comparison of the proposed methods to the K-means++ and K-means‖ methods is conducted using an extensive set of reference and synthetic large-scale datasets. Concerning the latter, a novel high-dimensional clustering data generation …
Improvements and applications of the elements of prototype-based clustering
2018
Clustering or cluster analysis is an essential part of data mining, machine learning, and pattern recognition. The most popularly applied clustering methods are partitioning-based or prototype-based methods. Prototype-based clustering methods usually have easy implementability and good scalability. These methods, such as K-means clustering, have been used for different applications in various fields. On the other hand, prototype-based clustering methods are typically sensitive to initialization, and the selection of the number of clusters for knowledge discovery purposes is not straightforward. In the era of big data, in high-velocity, ever-growing datasets, which can also be erroneous, outl…
Random utility approach to re-design passengers interchange nodes
2009
The passengers interchange node is a complex infrastructure, where the user can choose among different modal options to do his/her trip. These transport infrastructures have various terminals (car park, bus stop, metro, light rail, railway, extra-urban and sub urban bus and so on). The passengers, inside interchange node, have to reach these terminals walking often with luggage and for long distances. We calibrated a discrete choice model taking into account attributes able to explain the passenger behaviour by observed flows and generalized least squares technique. Furthermore an ex ante / ex post analysis was carried out in order to design and evaluate measure to incentive modal integrati…
On probabilistic interpretations of predicates
2016
In classical logic, any m-ary predicate is interpreted as an m-argument two-valued relation defined on a non-empty universe. In probability theory, m-ary predicates are interpreted as probability measures on the mth power of a probability space. m-ary probabilistic predicates are equivalently semantically characterized as m-dimensional cumulative distribution functions defined on Rm. The paper is mainly concerned with probabilistic interpretations of unary predicates in the algebra of cumulative distribution functions defined on R. This algebra, enriched with two constants, forms a bounded De Morgan algebra. Two logical systems based on the algebra of cumulative distributions are defined an…
Rationale and design of the DARWIN-T2D (DApagliflozin Real World evIdeNce in Type 2 Diabetes): A multicenter retrospective nationwide Italian study a…
2017
Background Randomized controlled trials (RCTs) in the field of diabetes have limitations inherent to the fact that design, setting, and patient characteristics may be poorly transferrable to clinical practice. Thus, evidence from studies using routinely accumulated clinical data are increasingly valued. Aims We herein describe rationale and design of the DARWIN-T2D (DApagliflozin Real World evIdeNce in Type 2 Diabetes), a multicenter retrospective nationwide study conducted at 50 specialist outpatient clinics in Italy and promoted by the Italian Diabetes Society. Data synthesis The primary objective of the study is to describe the baseline clinical characteristics (particularly HbA1c) of pa…
Managing for Results in Primary Education: Results of a Randomized Trial in Madagascar
2012
International audience; Using information from a randomized experiment carried out over the course of two school years in Madagascar, this paper evaluates the impact of specific actions designed to streamline and tighten the work processes of public primary school directors. The results show that interventions at the school level, reinforced by interventions at the district and subdistrict levels, succeeded overall in changing school heads' behavior toward better management. However, the average impact hides important heterogeneity. The impact of the intervention was significantly larger among school heads who had a nonpermanent contract and among school principals who were responsible for …
Nošķeltās vidējās vērtības pielietošana trūkstošu datu gadījumā randomizētos kontrolētos pētījumos
2022
Šajā darbā tiek apskatīta nošķeltā vidējā vērtība trūkstošu datu gadījumā un hipotēžu testi nošķelto vidējo vērtību salīdzināšanai -- permutāciju, Jūenas un EL tests. Tiek sniegts arī ieskats par daudzkārtējo imputāciju. Tiek izstrādāts simulāciju eksperiments, kur metodes tiek apskatītas praktiski un rezultāti tiek attēloti tabulās, lai tos būtu vieglāk salīdzināt. Tiek analizēts arī reāls datu piemērs. Darbā var novērot atšķirīgus rezultātus starp metodēm.
Meditāciju saistība ar smadzeņu darbību, izmantojot elektroencefalogrāfu (EEG), randomizēti kontrolētos pētījumos: metaanalīze
2020
Mūsdienās meditācijas tiek praktizētas ne tikai austrumos, bet kļuvušas populāras arī rietumu pasaulē. Šajā metaanalīzē tika meklēta atbilde uz jautājumu, kādas saistības pastāv starp meditācijām un galvas smadzeņu viļņu darbību pieaugušajiem randomizēti kontrolētos pētījumos. Metaanalīzē tika iekļauti trīs pētījumi ar alfas viļņu rādītājiem un kopējo izlases apjomu N = 94. Metaanalīzes aprēķini uzrāda nelielu statistiski nozīmīgu efektu r = 0,20 (p<0,001). Tādējādi šī pētījuma ietvaros secināms, ka pēc meditācijām novērojams neliels efekts uz galvas smadzeņu alfa viļņu rādītājiem. Tas saskan ar līdzšinējiem pētījumiem, kuros vairākkārt ir secināts, ka meditācijas spēj statistiski nozīmīgi …