scholarly journals One-Layer vs. Two-Layer SOM in the Context of Outlier Identification: A Simulation Study

2021 ◽  
Vol 11 (14) ◽  
pp. 6241
Author(s):  
Gabriel Antonio Valverde Castilla ◽  
José Manuel Mira McWilliams ◽  
Beatriz González-Pérez

In this work, we applied a stochastic simulation methodology to quantify the power of the detection of outlying mixture components of a stochastic model, when applying a reduced-dimension clustering technique such as Self-Organizing Maps (SOMs). The essential feature of SOMs, besides dimensional reduction into a discrete map, is the conservation of topology. In SOMs, two forms of learning are applied: competitive, by sequential allocation of sample observations to a winning node in the map, and cooperative, by the update of the weights of the winning node and its neighbors. By means of cooperative learning, the conservation of topology from the original data space to the reduced (typically 2D) map is achieved. Here, we compared the performance of one- and two-layer SOMs in the outlier representation task. The same stratified sampling was applied for both the one-layer and two-layer SOMs; although, stratification would only be relevant for the two-layer setting—to estimate the outlying mixture component detection power. Two distance measures between points in the map were defined to quantify the conservation of topology. The results of the experiment showed that the two-layer setting was more efficient in outlier detection while maintaining the basic properties of the SOM, which included adequately representing distances from the outlier component to the remaining ones.

2022 ◽  
Vol 14 (2) ◽  
pp. 915
Author(s):  
Klaudia Nowicka

All tangible and intangible elements of cultural heritage that the past has conceded to local communities create unique landscapes shaped by tightly connected anthropogenic and natural factors. This heritage is a keystone of local identity which plays a significant role in politics, economic development, society and world view. In some regions, such as in the Vistula delta in Poland, the cultural heritage has been created by consecutive groups of settlers who represented different values, beliefs and ways of life. On the one hand, such a rich heritage may be perceived as a valuable asset and become a landmark or tourism product of a region. On the other hand, it may be perceived as alien and unwanted by contemporary residents, especially when they are not descendants of the former communities. The main objective of the study presented herein is to analyse how the residents of the Vistula delta region, called Żuławy Wiślane, perceive and use cultural heritage of the Mennonites, representing the most extraordinary group of settlers who used to live in the region. The analysis covers original data gathered during survey research in the period of 2017–2018 under the project Miniatura I “Perception and usage of cultural heritage of the Vistula delta Mennonites” financed by the National Science Centre in Poland.


Author(s):  
Juergen Perl

Processes in sport like motions or games are influenced by communication, interaction, adaptation, and spontaneous decisions. Therefore, on the one hand, those processes are often fuzzy and unpredictable and so have not extensively been dealt with, yet. On the other hand, most of those processes structurally are roughly determined by intention, rules, and context conditions and so can be classified by means of information patterns deduced from data models of the processes. Self organizing neural networks of type Kohonen Feature Map (KFM) help for classifying information patterns – either by mapping whole processes to corresponding neurons (see Perl & Lames, 2000; McGarry & Perl, 2004) or by mapping process steps to neurons, which then can be connected by trajectories that can be taken as process patterns for further analyses (see examples below). In any case, the dimension of the original data (i.e. the number of contained attributes) is reduced to the dimension of the representing neuron (normally 2 or 3), which makes it much easier to deal with. Additionally, extensions of the KFM-approach are introduced, which are able to flexibly adjust the net to dynamically changing training situations. Moreover, those extensions allow for simulating adaptation processes like learning or tactical behaviour. Finally, a current project is introduced, where tactical processes in soccer are analysed under the aspect of simulation-based optimization.


2020 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Mazen Hassan

Purpose This paper aims to examine why the alliance formed between non-Islamist forces and state actors to oust Mohamed Morsi from power in 2013 broke down quickly. Design/methodology/approach This paper makes use of original data set derived from three waves of surveys fielded in 2011, 2014 and 2015 that ask questions about public threat perception. Around 10 elite interviews were also conducted to further test the study’s hypothesis. Findings On the one hand, non-Islamists, civic forces challenged the status and interests of state actors in a way that made state actors view them with heightened distrust. On the other, many civic forces, in face of high threat perception, prioritized law during and order after Morsi’s removal, driven – at least partly – by shifts in public attitudes. Originality/value Many views look at transitions in the Arab world from the angle of how Islamists interact with traditional power holders. Such an approach, however, could be reductionist in many ways because it disregards civic forces. This is a point this paper seeks to address.


2004 ◽  
Vol 21 (2) ◽  
pp. 121-125 ◽  
Author(s):  
Akihiko Ibukiyama

AbstractWe derive age–metallicity relations (AMRs) and orbits for the 1658 solar neighbourhood stars for which accurate distances are measured by the Hipparcos satellite. The sample comprises 1382 thin disk stars, 229 thick disk stars, and 47 halo stars according to their orbital parameters. We find a considerable scatter for thin disk AMRs along the one-zone Galactic chemical evolution (GCE) model. Orbits and metallicities of thin disk stars show no clear relation to each other. The scatter along the AMR exists even if stars with the same orbits are selected. We examine simple extensions of one-zone GCE models which account for inhomogeneity in the effective yield and inhomogeneous star formation rate in the Galaxy. Both extensions of the one-zone GCE model cannot account for the scatter in the age–[Fe/H]–[Ca/Fe] relation simultaneously. We conclude, therefore, that the scatter along the thin disk AMR is an essential feature in the formation and evolution of the Galaxy. The AMR for thick disk stars shows that star formation terminated 8 Gyr ago in the thick disk. As previously reported, thick disk stars are more Ca-rich than thin disk stars with the same [Fe/H]. We find that thick disk stars show a vertical abundance gradient. These three facts — AMR, vertical gradient, and [Ca/Fe]–[Fe/H] relation — support monolithic collapse and/or accretion of satellite dwarf galaxies as likely thick disk formation scenarios.


Geophysics ◽  
2019 ◽  
Vol 84 (4) ◽  
pp. Q27-Q36 ◽  
Author(s):  
Lele Zhang ◽  
Jan Thorbecke ◽  
Kees Wapenaar ◽  
Evert Slob

We have developed a scheme that retrieves primary reflections in the two-way traveltime domain by filtering the data. The data have their own filter that removes internal multiple reflections, whereas the amplitudes of the retrieved primary reflections are compensated for two-way transmission losses. Application of the filter does not require any model information. It consists of convolutions and correlations of the data with itself. A truncation in the time domain is applied after each convolution or correlation. The retrieved data set can be used as the input to construct a better velocity model than the one that would be obtained by working directly with the original data and to construct an enhanced subsurface image. Two 2D numerical examples indicate the effectiveness of the method. We have studied bandwidth limitations by analyzing the effects of a thin layer. The presence of refracted and scattered waves is a known limitation of the method, and we studied it as well. Our analysis indicates that a thin layer is treated as a more complicated reflector, and internal multiple reflections related to the thin layer are properly removed. We found that the presence of refracted and scattered waves generates artifacts in the retrieved data.


Author(s):  
Serhii Oksak

Abstract. Problem. The temperature of mastic as-phalts mixing is significantly higher than the one for hot asphalt mixtures. It is due using of high viscous bitumens as a mixture component and realizing the main advantage of mastic asphalt – its ability to flow and compact with its own weight at high tempera-ture. There is no uniform way to assign the process temperatures for mastic asphalt. Goal. To find the process temperatures for mastic asphalt with bitumen binders differ by viscosity as a researching on tem-perature influence and influence of binder concentra-tion on properties of mastic asphalt is a goal of this research work. Methodology. The mastic asphalt mixtures and asphalt concrete with conventional pavement bitumen 70/100 (penetration is 85 dmm) and high viscous bitumen (penetration is 38 dmm) are chosen for research. The mineral aggregates contain 42 % of crushed granite 5-10 mm, 33 % of granite screening and 33 % of limestone powder. The properties of mastic asphalts are evaluated with EN 12697-20:2003, SOU 42.1-37641918-106:2013 and DSTU EN 13108-6:2018. Results. It was found a significant influence of amount and viscosity of bitu-men and a temperature on workability of mastic as-phalt mixture and on properties of mastic asphalt concrete. Originality. The comparative analysis of workability of mastic asphalt mixture and properties of mastic asphalt concrete with bitumens differ in viscosity is a scientific originality of presented re-search. Practical value. Due to research results it is not reasonable to assign the temperature that sup-plies viscosity of 0.5 Pa•s as a temperature of mixing of mastic asphalt. The temperature conditions of mastic asphalt mixing are much higher (200 – 240 °C). To evaluate workability and find a minimal acceptable temperature of mixture it is possible to use the thickness of the mixture layer after spreading as an indicator. With this the acceptable workability of the mastic asphalt mixture can be characterized by thickness of its layer 30 mm or less after spreading.


Author(s):  
Dragana Popovic Renella ◽  
Vojin Senk ◽  
Fuada Stankovic

This is a study of the process of the development of high-tech start-up companies through the mechanisms of bootstrapping in two extremely different environments: the one of highly industrialized countries, such as USA and Switzerland; the other of Serbia, which is a post-communist transition country with particular difficulties. The research method is the analysis of case studies. One case study of US and two of Swiss start-ups build the base for the analysis. All three cases show common patterns: from the very beginning, these companies sell R&D services in their fields of expertise; and using the cash from these early sales, and the information feedback from cooperation with the early customers, they develop their resources and, eventually, also their own high-tech products. The essential feature of this process is the selling of R&D services and the first products in the neighborhoods. Then also two cases of high-tech start-ups from Serbia are analyzed. Both Serbian start-ups are founded in partnership with small high-tech companies from highly industrialized countries (Switzerland and Germany). The Western partners use their reputations and contacts to enable the early sales of the Serbian start-ups in the industrialized countries. This is crucial for the Serbian start-ups, because they have no domestic market for R&D services. Apart of this element, all other essential patterns of the Serbian cases are very similar to those of the Western cases.


Gragoatá ◽  
2021 ◽  
Vol 26 (54) ◽  
pp. 436-444
Author(s):  
Britta Schneider ◽  
Konstanze Jungbluth

Data from plurilingual Belize shows that not everyone recognizes stability as an essential feature of aspoken language. Belizeans consider the use of Kriol as a symbol of belonging but foreground its readiness for variation across communities in space and time. Their use of liquid languages is a different form of cultural construction than the one our textbooks show. It questions a good part of linguistics and reveals its possibly Eurocentric point of view.


2020 ◽  
Author(s):  
Christopher Reinders Folmer ◽  
Malouke Kuiper ◽  
Elke Olthuis ◽  
Emmeke Barbara Kooistra ◽  
Anne Leonre de Bruijn ◽  
...  

In the month of May, the Netherlands moved out of the “intelligent lockdown”, and into the “1.5 meter society”, which aims to mitigate the COVID-19 pandemic by means of safe-distance measures. This paper assesses how Dutch citizens have complied with these social distancing measures. It analyses data from two surveys conducted in May (between 8-14 and between 22-26) among nationally representative samples (N = 984 and N = 1021). We find that a combination of factors explain social distancing compliance. On the one hand we see that people are more likely to comply if they have an intrinsic motivation to do so, when they have the capacity to comply, when they have good impulse control, when they think compliance is normal, and when they see a general duty to obey rules generally. The paper also assesses how compliance has changed over time, assessing changes in May as well as how these are different from compliance with lockdown measures in April. During this period, there has been a gradual decline in compliance that coincides with a decline in intrinsic motivations and capacity for compliance, and there has been an increase in opportunities to violate the measures. The paper assesses what these changes may mean for current and future success of Covid-19 mitigation measures.


2008 ◽  
pp. 371-380
Author(s):  
Takao Ito

One of the most important issues in data mining is to discover an implicit relationship between words in a large corpus and labels in a large database. The relationship between words and labels often is expressed as a function of distance measures. An effective measure would be useful not only for getting the high precision of data mining, but also for time saving of the operation in data mining. In previous research, many measures for calculating the one-to-many relationship have been proposed, such as the complementary similarity measure, the mutual information, and the phi coefficient. Some research showed that the complementary similarity measure is the most effective. The author reviewed previous research related to the measures in one-to-many relationships and proposed a new idea to get an effective one, based on the heuristic approach in this article.


Sign in / Sign up

Export Citation Format

Share Document