Graph Mining Based SOM: A Tool to Analyze Economic Stability

Resta, Marina

doi:10.5772/51240

Cited by 4 publications

(3 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Each neuron is fully connected to all the source nodes in the input layer, and the connection weights are initialized with small random values, or with appropriate input values. Training a SOM requires a number of iterative steps (Resta, 2012): (1) evaluate the distance between x and the vector of weights of the synaptic connections entering in each neuron; (2) select the neuron (node) with the smallest distance to x (i.e., "winner neuron" or Best Matching Unit -BMU); (3) correct the position (i.e., by modifying the weights) of each node according to the results of Step 2, in order to preserve the network topology. This iterative process continues until a stopping criterion is reached.…”

Section: Unsupervised Ml: Self-organizing Mapsmentioning

confidence: 99%

Fighting Over-Indebtedness: An Artificial Intelligence Approach: An Abstract

Ferreira

Pinto

Herter

et al. 2022

Developments in Marketing Science: Proceedings of the Academy of Marketing Science

View full text Add to dashboard Cite

This paper examines how artificial intelligence may contribute to better understanding and overcome over-indebtedness in contexts of severe economic austerity. We analyze a field database of 1,654 over-indebted households with a high risk of poverty. Artificial intelligence algorithms are used to identify distinguishable over-indebtedness clusters and to predict over-indebtedness risk factors within each cluster. First, unsupervised machine learning using Self-Organizing Maps generated three over-indebtedness clusters: low-income families (31.27%), low credit control families (37.40%), and families affected by abrupt economic crisis (31.33%). Second, supervised machine learning with exhaustive grid search hyperparameters (32,730 predictive models) suggest that Nu-Support Vector Machine had the best accuracy in predicting families' over-indebtedness risk factors (89.5%). These findings extend previous research by proposing a multifaced and yet organized bottom-up approach to over-indebtedness and poverty risk.

show abstract

Section: Unsupervised Ml: Self-organizing Mapsmentioning

confidence: 99%

Fighting Over-Indebtedness: An Artificial Intelligence Approach: An Abstract

Ferreira

Pinto

Herter

et al. 2022

Developments in Marketing Science: Proceedings of the Academy of Marketing Science

View full text Add to dashboard Cite

show abstract

“…SOMs use unsupervised training algorithms and belong to a general class of ANNs based on nonlinear regression techniques that can be trained to organize data so as to disclose unknown patterns or structures (Deboeck and Kohonen, 1988). SOMs have been used in order to make visual predictions of different phenomena, but only recently in economic studies (Sarlin and Peltonen, 2013;Resta, 2012;Lu and Wang, 2010;Marghescu et al, 2010;Arciniegas-Rueda and Arciniegas, 2009;Eklund et al, 2008). In this study we make use of SOMs to analyse experts' expectations about the state of the economy in fourteen European countries before and after the financial crisis of 2008.…”

Section: Introductionmentioning

confidence: 99%

A self-organizing map analysis of survey-based agents׳ expectations before impending shocks for model selection: The case of the 2008 financial crisis

Clavería

Monte

Porras

2016

International Economics

View full text Add to dashboard Cite

A self-organizing map analysis of survey-based agents' expectations before impending shocks for model selection: The case of the 2008 financial crisis ABSTRACT By means of Self-Organizing Maps we cluster fourteen European countries according to the most suitable way to model their agents' expectations. Using the financial crisis of 2008 as a benchmark, we distinguish between those countries that show a progressive anticipation of the crisis and those where sudden changes in expectations occur. We compare the forecasting performance of Artificial Neural Networks to that of three different time series models and find that neural networks outperform time series models when there are brisk changes in expectations before impending shocks. Conversely, in countries where expectations show a smooth transition towards recession, ARIMA models outperform neural networks. This result demonstrates the usefulness of clustering techniques for selecting the most appropriate method to forecast expectations according to their behaviour.

show abstract

“…SOMs use unsupervised training algorithms and belong to a general class of ANNs based on nonlinear regression techniques that can be trained to organize data so as to disclose so far unknown patterns or structures (Deboeck and Kohonen, 1988). SOMs have been used in order to make visual predictions of different phenomena, but only recently in economic studies (Sarlin and Peltonen, 2013;Resta, 2012;Lu and Wang, 2010;Marghescu et al, 2010;Arciniegas-Rueda and Arciniegas, 2009;Eklund et al, 2008). In this study we make use of SOMs to analyze experts' expectations about the state of the economy in fourteen European countries before and after the financial crisis of 2008.…”

Section: Introductionmentioning

confidence: 99%

Self-Organizing Map Analysis of Agents' Expectations. Different Patterns of Anticipation of the 2008 Financial Crisis

2015

View full text Add to dashboard Cite

By means of Self-Organizing Maps we cluster fourteen European countries according to the most suitable way to model their agents' expectations. Using the financial crisis of 2008 as a benchmark, we distinguish between those countries that show a progressive anticipation of the crisis and those where sudden changes in expectations occur. By mapping the trajectory of economic experts' expectations prior to the recession we find that when there are brisk changes in expectations before impending shocks, Artificial Neural Networks are more suitable than time series models for modelling expectations. Conversely, in countries where expectations show a smooth transition towards recession, ARIMA models show the best forecasting performance. This result demonstrates the usefulness of clustering techniques for selecting the most appropriate method to model and forecast expectations according to their behaviour..

show abstract

Graph Mining Based SOM: A Tool to Analyze Economic Stability

Cited by 4 publications

References 23 publications

Fighting Over-Indebtedness: An Artificial Intelligence Approach: An Abstract

Fighting Over-Indebtedness: An Artificial Intelligence Approach: An Abstract

A self-organizing map analysis of survey-based agents׳ expectations before impending shocks for model selection: The case of the 2008 financial crisis

Self-Organizing Map Analysis of Agents' Expectations. Different Patterns of Anticipation of the 2008 Financial Crisis

Contact Info

Product

Resources

About