scholarly journals Classifying non-banking financial institutions based on their financial performance

2019 ◽  
Vol 1 (1) ◽  
pp. 185-193
Author(s):  
Adrian Costea
Keyword(s):  
Financial Institutions ◽  
Capital Adequacy ◽  
The Self ◽  
Three Dimensions ◽  
Self Organizing Map ◽  
Two Dimensional ◽  
Self Organizing Maps ◽  
Performance Dimensions ◽  
Self Organising Map ◽  
Self Organizing

Abstract In this paper we evaluate comparatively the performance of non-banking financial institutions in Romania by the means of unsupervised neural networks in terms of Kohonen’ Self-Organizing Maps algorithm. We create a benchmarking model in the form of a two-dimensional map (a self-organizing map) that can be used to assess visually the performance of non-banking financial institutions based on different performance dimensions, such as capital adequacy, assets’ quality and profitability. We use the following indicators: Equity ratio (Leverage) for the capital adequacy dimension, Loans granted to clients (net value) / total assets (net value) for the assets’ quality dimension and Return on assets (ROA) for the profitability dimension. We have excluded from our analysis the other three dimensions used in evaluating the performance of banks, due to lack of data (for the two qualitative dimensions: quality of ownership and management) and irrelevance with the NFIs’ sector (liquidity). The proposed model is based on the Self-Organising Map algorithm which creates a two-dimensional map (e.g. 6x4 = 24 neurons) from p-dimensional input data. The data were collected for eleven non-banking financial institutions for four years 2007-2010, in total 44 observations. Using the visualization capabilities of the Self-Organising Map model and the trajectories we show the movements of the three non-banking financial institutions with the worst performance: the largest underperformer denoted with X, the second largest underperformer denoted with Y and the third largest underperformer denoted with Z between 2007 and 2010.

scholarly journals On building early-warning systems for preventing the deterioration of financial institutions’ performance

2019 ◽  
Vol 1 (1) ◽  
pp. 194-202
Author(s):  
Adrian Costea
Keyword(s):  
Financial Institutions ◽  
Early Warning ◽  
Warning System ◽  
Early Warning Systems ◽  
Capital Adequacy ◽  
Self Organizing Map ◽  
Self Organizing Maps ◽  
Performance Dimensions ◽  
Som Model ◽  
Self Organizing

Abstract This paper assesses the financial performance of Romania’s non-banking financial institutions (NFIs) using a neural network training algorithm proposed by Kohonen, namely the Self-Organizing Maps algorithm. The algorithm takes the financial dataset and positiones each observation into a self-organizing map (a two-dimensional map) which can be latter used to visualize the trajectories of an individual NFI and explain it based on different performance dimensions, such as capital adequacy, assets’ quality and profitability. Further, we use the map as an early-warning system that would accurately forecast the NFIs future performance (whether they would stay or be eliminated from the NFI’s Special Register three quarters into the future). The results are promising: the model is able to correctly predict NFIs’ performance movements. Finally, we compared the results of our SOM-based model with those obtained by applying a multivariate logit-based model. The SOM model performed worse in discriminating the NFIs’ performance: the performance classes were not clearly defined and the model lacked the interpretability of the results. In the contrary, the multivariate logit coefficients have nice interpretability and an individual default probability estimate is obtained for each new observation. However, we can benefit from the results of both techniques: the visualization capabilities of the SOM model and the interpretability of multivariate logit-based model.

The self-organizing map in synoptic climatological research

2011 ◽  
Vol 35 (1) ◽  
pp. 109-119 ◽  
Author(s):  
Scott C. Sheridan ◽  
Cameron C. Lee
Keyword(s):  
Spatial Patterns ◽  
The Self ◽  
Synoptic Climatology ◽  
Self Organizing Map ◽  
Two Dimensional ◽  
Clustering Methods ◽  
Traditional Methods ◽  
Self Organizing Maps ◽  
Dimensional Array ◽  
Self Organizing

Self-organizing maps (SOMs) are a relative newcomer to synoptic climatology; the method itself has only been utilized in the field for around a decade. In this article, we review the major developments and climatological applications of SOMs in the literature. The SOM can be used in synoptic climatological analysis in a manner similar to most other clustering methods. However, as the results from a SOM are generally represented by a two-dimensional array of cluster types that ‘self-organize’, the synoptic categories in the array effectively represent a continuum of synoptic categorizations, compared with discrete realizations produced through most traditional methods. Thus, a larger number of patterns can be more readily understood, and patterns, as well as transitional nodes between patterns, can be discerned. Perhaps the most intriguing development with SOMs has been the new avenues of visualization; the resultant spatial patterns of any variable can be more readily understood when displayed in a SOM. This improved visualization has led to SOMs becoming an increasingly popular tool in various research with climatological applications from other disciplines as well.

scholarly journals The Spread of the COVID-19 Outbreak in Brazil: An Overview by Kohonen Self-Organizing Map Networks

Medicina ◽  
2021 ◽  
Vol 57 (3) ◽  
pp. 235
Author(s):  
Diego Galvan ◽  
Luciane Effting ◽  
Hágata Cremasco ◽  
Carlos Adam Conte-Junior
Keyword(s):  
The Self ◽  
Self Organizing Map ◽  
Virus Spread ◽  
Self Organizing Maps ◽  
The North ◽  
Som Clustering ◽  
Unsupervised Neural Networks ◽  
Mining Tools ◽  
Temporal Spread ◽  
Self Organizing

Background and objective: In the current pandemic scenario, data mining tools are fundamental to evaluate the measures adopted to contain the spread of COVID-19. In this study, unsupervised neural networks of the Self-Organizing Maps (SOM) type were used to assess the spatial and temporal spread of COVID-19 in Brazil, according to the number of cases and deaths in regions, states, and cities. Materials and methods: The SOM applied in this context does not evaluate which measures applied have helped contain the spread of the disease, but these datasets represent the repercussions of the country’s measures, which were implemented to contain the virus’ spread. Results: This approach demonstrated that the spread of the disease in Brazil does not have a standard behavior, changing according to the region, state, or city. The analyses showed that cities and states in the north and northeast regions of the country were the most affected by the disease, with the highest number of cases and deaths registered per 100,000 inhabitants. Conclusions: The SOM clustering was able to spatially group cities, states, and regions according to their coronavirus cases, with similar behavior. Thus, it is possible to benefit from the use of similar strategies to deal with the virus’ spread in these cities, states, and regions.

Modeling and Production of Robot Trajectories Using the Temporal Parametrized Self Organizing Maps

2003 ◽  
Vol 13 (02) ◽  
pp. 119-127 ◽  
Author(s):  
Antonio Carlos Padoan ◽  
Guilherme de A. Barreto ◽  
Aluizio F. R. Araújo
Keyword(s):  
Inverse Kinematics ◽  
The Self ◽  
Self Organizing Map ◽  
Self Organizing Maps ◽  
Neural Architecture ◽  
New States ◽  
Simulation Results ◽  
Self Organizing

In this paper we proposed an unsupervised neural architecture, called Temporal Parametrized Self Organizing Map (TEPSOM), capable of learning and reproducing complex robot trajectories and interpolating new states between the learned ones. The TEPSOM combines the Self-Organizing NARX (SONARX) network, responsible for coding the temporal associations of the robotic trajectory, with the Parametrized Self-Organizing (PSOM) network, responsible for an efficient interpolation mechanism acting on the SONARX neurons. The TEPSOM network is used to model the inverse kinematics of the PUMA 560 robot during the execution of trajectories with repeated states. Simulation results show that the TEPSOM is more accurate than the SONARX in the reproduction of the learned trajectories.

Using the Self-Organizing Map as a Visualization Tool in Financial Benchmarking

2003 ◽  
Vol 2 (3) ◽  
pp. 171-181 ◽  
Author(s):  
Tomas Eklund ◽  
Barbro Back ◽  
Hannu Vanharanta ◽  
Ari Visa
Keyword(s):  
Performance Analysis ◽  
Pulp And Paper ◽  
The Self ◽  
Financial Data ◽  
Visualization Tool ◽  
Self Organizing Map ◽  
Self Organizing Maps ◽  
Time Period ◽  
The World ◽  
Self Organizing

In this paper, we illustrate the use of the self-organizing map technique for financial performance analysis and benchmarking. We build a database of financial ratios indicating the performance of 91 international pulp and paper companies for the time period 1995–2001. We then use the self-organizing map technique to analyze and benchmark the performance of the five largest pulp and paper companies in the world. The results of the study indicate that by using the self-organizing maps, we are able to structure, analyze, and visualize large amounts of multidimensional financial data in a meaningful manner.

scholarly journals Village Grouping In Southwest Maluku District Based On Poverty Characteristics Using Self Organizing Maps (SOM) Methods

2020 ◽  
Vol 5 (1) ◽  
pp. 16-20
Author(s):  
Monalisa E. Rijoly ◽  
F. L. Lumalessil ◽  
B. P. Tomasouw
Keyword(s):  
Standard Deviation ◽  
Provincial Government ◽  
The Self ◽  
Self Organizing Map ◽  
Self Organizing Maps ◽  
Network Method ◽  
The Government ◽  
Deviation Ratio ◽  
Standard Deviation Ratio ◽  
Self Organizing

Poverty is one of the fundamental problems that has become the center of attention of the Maluku Provincial government, especially Southwest Maluku Regency. This study aims to provide information to the government about village grouping based on poverty characteristics in Southwest Maluku Regency using the Self Organizing Map network method. In this network, a layer containing neurons will arrange itself based on the input of a certain value in a group known as a cluster. In the grouping process, 3 results were obtained with the best grouping II results because they had the smallest standard deviation ratio value.

scholarly journals Using Self-Organizing Maps for Rural Territorial Typology

2011 ◽  
pp. 107-126 ◽  
Author(s):  
Marcos Santos da Silva ◽  
Edmar Ramos de Siqueira ◽  
Olívio Teixeira ◽  
Maria Manos ◽  
Antônio Monteiro
Keyword(s):  
Neural Network ◽  
Temporal Analysis ◽  
The Self ◽  
Self Organizing Map ◽  
Clustering Methods ◽  
Self Organizing Maps ◽  
Policy Makers ◽  
Spatial Temporal Analysis ◽  
Self Organizing

This work assessed the capacity of the self-organizing map, an unsupervised artificial neural network, to aid the process of territorial design through visualization and clustering methods applied to a multivariate geospatial temporal dataset. The method was applied in the case study of Sergipe‘s institutional regional partition (Territories of Identity). Results have shown that the proposed method can improve the exploratory spatial-temporal analysis capacity of policy makers that are interested in territorial typology. A new partition for rural planning was elaborated and confirmed the coherence of the Territories of Identity.

scholarly journals Saviorganizuojančių neuroninių tinklų sistemų lyginamoji analizė

2009 ◽  
Vol 50 ◽  
pp. 334-339
Author(s):  
Pavel Stefanovič ◽  
Olga Kurasova
Keyword(s):  
Data Clustering ◽  
The Self ◽  
Data Sets ◽  
Self Organizing Map ◽  
Self Organizing Maps ◽  
Learning Rules ◽  
Main Target ◽  
Similarities And Differences ◽  
Graphical Presentation ◽  
Self Organizing

Straipsnyje nagrinėjamos ir lyginamos tarpusavyje trys saviorganizuojančių neuroninių tinklų (SOM) sistemos: NeNet, SOM-Toolbox ir Databionic ESOM. Pagrindinis šių sistemų tikslas yra suskirstyti duomenis į klasterius pagal jų panašumą, pateikti juos SOM žemėlapyje. Sistemos viena nuo kitos skiriasi duomenų pateikimu, mokymo taisyklėmis, vizualizavimo galimybėmis, todėl čia aptariami sistemų panašumai ir skirtumai. SOM žemėlapiams mokyti ir vizualizuoti naudojami irisų ir stikloduomenys.Comparative Analysis of Self-Organizing Map SystemsPavel Stefanovič, Olga Kurasova SummaryIn the article, we compare three systems of self-organizing maps: NeNet, SOM-Toolbox and Databionic ESOM. The main target of the usage of the systems is data clustering and their graphical presentation on the self-organizing map (SOM). The self-organizing maps are one of types of artifi cial neural networks. The SOM systems are different one from other in their interfaces, the data pre-processing, learning rules, visualization manners, etc. Similarities and differences of the systems have been highlighted here. The experiments have been carried out with two data sets: iris and glass. Quantization and topographic errors of SOMs have been estimated, too.an>

scholarly journals Tekstinių dokumentų panašumų paieška naudojant saviorganizuojančius neuroninius tinklus ir k vidurkių metodą

2013 ◽  
Vol 65 ◽  
pp. 24-33
Author(s):  
Pavel Stefanovič ◽  
Olga Kurasova
Keyword(s):  
The Self ◽  
Self Organizing Map ◽  
Text Documents ◽  
Self Organizing Maps ◽  
Text Data ◽  
Document Database ◽  
Cluster A ◽  
The Republic ◽  
Sum Of Distances ◽  
Self Organizing

Straipsnyje nagrinėjama dokumentų panašumų paieška naudojant du populiarius metodus: saviorganizuojančius neuroninius tinklus (SOM) ir k vidurkių metodą. Vienas iš šių metodų tikslų – suskirstyti duomenis į klasterius pagal jų panašumą. Analizuota tekstinių dokumentų matricos sudarymo faktorių įtaka gautiems rezultatams. SOM kokybei įvertinti pasiūlyti du nauji matai, skirti klasifi kuotiems duomenims, kurių reikšmės parodo susidariusių klasterių išsidėstymą SOM žemėlapyje. Pirmasis matas parodo, kaip gerai tos pačios klasės duomenys išsidėsto žemėlapyje vienas šalia kito, antrasis matas – kaip toli yra skirtingų klasių centrai. K vidurkių metodu gautų rezultatų kokybei įvertinti skaičiuota suma nuo klasterio centro iki klasterio narių bei įvertintas klasių nesutapimas su klasteriais. Eksperimentiniams tyrimams atlikti pasirinkti tekstiniai dokumentai, paimti iš Lietuvos Respublikos Seimo dokumentų bazės.Similarity analysis of text documents by self-organizing maps and k-means Pavel Stefanovič, Olga Kurasova SummaryIn this paper, we try to fi nd similarities of different text documents by the self-organizing map (SOM) and k-means method. One of the main goals of these methods is to cluster a dataset. Using SOM, the similarities of documents can be observed visually. Both methods can be used only for numerical information, so we analyse the different options by converting text data on to numerical in order to get better results. To estimate the SOM quality, when the classifi ed data are analysed, we propose two new measures: distances between SOM cells, corresponding to data items assigned to the same class, and the distance between centres of SOM cells, corresponding to different classes. We also analyse the results of visualization by self-organizing maps. In order to estimate the k-means quality, we calculate the sum of distances between cluster centres and class members and also we estimate assignment of the data from particular classes to the clusters. The experiments have been carried out using three datasets ocquired from the document database of Seimas of the Republic of Lithuania.font-family: Calibri, sans-serif;"> 

scholarly journals Deep Self-Organizing Map of Convolutional Layers for Clustering and Visualizing Image Data

2021 ◽  
Vol 3 (4) ◽  
pp. 879-899
Author(s):  
Christos Ferles ◽  
Yannis Papanikolaou ◽  
Stylianos P. Savaidis ◽  
Stelios A. Mitilineos
Keyword(s):  
Neural Networks ◽  
Image Data ◽  
Building Blocks ◽  
The Self ◽  
Self Organizing Map ◽  
Self Organizing Maps ◽  
Deep Architecture ◽  
Unsupervised Deep Learning ◽  
Deep Learning Model ◽  
Self Organizing

The self-organizing convolutional map (SOCOM) hybridizes convolutional neural networks, self-organizing maps, and gradient backpropagation optimization into a novel integrated unsupervised deep learning model. SOCOM structurally combines, architecturally stacks, and algorithmically fuses its deep/unsupervised learning components. The higher-level representations produced by its underlying convolutional deep architecture are embedded in its topologically ordered neural map output. The ensuing unsupervised clustering and visualization operations reflect the model’s degree of synergy between its building blocks and synopsize its range of applications. Clustering results are reported on the STL-10 benchmark dataset coupled with the devised neural map visualizations. The series of conducted experiments utilize a deep VGG-based SOCOM model.

Sign in / Sign up

Export Citation Format

Share Document