scholarly journals Business Client Segmentation in Banking Using Self-Organizing Maps

2014 ◽  
Vol 8 (2) ◽  
pp. 32-41 ◽  
Author(s):  
Mirjana Pejić Bach ◽  
Sandro Juković ◽  
Ksenija Dumičić ◽  
Nataša Šarlija
Keyword(s):  
Foreign Trade ◽  
Selection Criteria ◽  
Clustering Algorithm ◽  
Annual Income ◽  
Clustering Methods ◽  
Self Organizing Maps ◽  
Marketing Activities ◽  
Ward Clustering ◽  
Statistical Clustering ◽  
Self Organizing

Abstract Segmentation in banking for the business client market is traditionally based on size measured in terms of income and the number of employees, and on statistical clustering methods (e.g. hierarchical clustering, k-means). The goal of the paper is to demonstrate that self-organizing maps (SOM) effectively extend the pool of possible criteria for segmentation of the business client market with more relevant criteria, including behavioral, demographic, personal, operational, situational, and cross-selling products. In order to attain the goal of the paper, the dataset on business clients of several banks in Croatia, which, besides size, incorporates a number of different criteria, is analyzed using the SOM-Ward clustering algorithm of Viscovery SOMine software. The SOM-Ward algorithm extracted three segments that differ with respect to the attributes of foreign trade operations (import/export), annual income, origin of capital, important bank selection criteria, views on the loan selection and the industry. The analyzed segments can be used by banks for deciding on the direction of further marketing activities.

Assessment of Clusteranalysis and Self-Organizing Maps

1998 ◽  
Vol 06 (02) ◽  
pp. 139-149 ◽  
Author(s):  
Helge Petersohn
Keyword(s):  
Neural Networks ◽  
Market Segmentation ◽  
Clustering Methods ◽  
Clustering Method ◽  
Self Organizing Maps ◽  
Methodical Approach ◽  
Starting Conditions ◽  
Marketing Activities ◽  
Special Approach ◽  
Self Organizing

Market segmentation represents a central problem of preparing marketing activities. The methodical approach of this problem is supported by clustering methods. Available data are used to detect common grounds regarding their quality structures. Therefore statistics provides various methods for cluster analysis. Self-organizing maps are another possibility to form classes. They are a special approach of the artificial neural networks. The statistical methods and these methods, which are based on organic processes of our brain, offer different solutions although the starting conditions are the same. Often decisions about investigations are based on such solutions. Therefore the results of clustering are very important to reveal systematic information about the size of classes and their structure. Methodical notes are needed for the use of any clustering method. This paper offers a simplified way to select the best result for clustering.

DGFCM: A New Dynamic Clustering Algorithm

2014 ◽  
Vol 556-562 ◽  
pp. 3945-3948
Author(s):  
Xin Qing Geng ◽  
Hong Yan Yang ◽  
Feng Mei Tao
Keyword(s):  
Vector Space ◽  
Fuzzy Clustering ◽  
Clustering Algorithm ◽  
Vector Space Model ◽  
Dynamic Clustering ◽  
Self Organizing Maps ◽  
Space Model ◽  
Fuzzy Clustering Algorithm ◽  
Present Algorithm ◽  
Self Organizing

This paper applies the dynamic self-organizing maps algorithm to determining the number of clustering. The text eigenvector is acquired based on the vector space model (VSM) and TF.IDF method. The number of clustering acquired by the dynamic self-organizing maps. The threshold GT control the network’s growth.Compared to the traditional fuzzy clustering algorithm, the present algorithm possesses higher precision. The example demonstrates the effectiveness of the present algorithm.

Codebook Clustering by Self-Organizing Maps for Fractal Image Compression

Fractals ◽  
1997 ◽  
Vol 05 (supp01) ◽  
pp. 27-38 ◽  
Author(s):  
Raouf Hamzaoui
Keyword(s):  
Image Compression ◽  
Clustering Algorithm ◽  
Not Given ◽  
Fractal Image Compression ◽  
Self Organizing Maps ◽  
Classification Schemes ◽  
Fast Encoding ◽  
Fractal Image ◽  
Traditional Classification ◽  
Self Organizing

A fast encoding scheme for fractal image compression is presented. The method uses a clustering algorithm based on Kohonen's self-organizing maps. Domain blocks are clustered, yielding a classification with a notion of distance which is not given in traditional classification schemes.

scholarly journals Weighted Particle Swarm Clustering Algorithm for Self-Organizing Maps

2020 ◽  
Vol 2 (2) ◽  
pp. 85-95
Author(s):  
Guorong Cui ◽  
Hao Li ◽  
Yachuan Zhang ◽  
Rongjing Bu ◽  
Yan Kang ◽  
...  
Keyword(s):  
Clustering Algorithm ◽  
Particle Swarm ◽  
Self Organizing Maps ◽  
Self Organizing

scholarly journals Characterization of Regional Wind Patterns Using Self-Organizing Maps: Impact on Dallas–Fort Worth Long-Term Ozone Trends

2019 ◽  
Vol 58 (4) ◽  
pp. 757-772 ◽  
Author(s):  
Alexander Kotsakis ◽  
Yunsoo Choi ◽  
Amir H. Souri ◽  
Wonbae Jeon ◽  
James Flynn
Keyword(s):  
Clustering Algorithm ◽  
Regional Scale ◽  
Ozone Production ◽  
Barnett Shale ◽  
Fort Worth ◽  
Pressure System ◽  
Self Organizing Maps ◽  
The Difference ◽  
Self Organizing ◽  
Wind Patterns

AbstractThis study analyzes wind patterns in the Dallas–Fort Worth (DFW) area to gain a clearer understanding of meteorological patterns that have historically led to ozone exceedances in this region. Using a clustering algorithm called “self-organizing maps,” we analyzed five notable characteristic regional wind patterns that occurred between April and October in 2000–14. A regional-scale high pressure system, cluster 2, produced weak southeast winds over DFW and accounted for 35.2% of ozone exceedances. Clusters 1 and 5, characterized by southwesterly winds over the DFW area, were together associated with one-third of total ozone exceedances and show quantifiable impacts of the Barnett Shale region on downwind ozone production. Cluster 3, associated with Bermuda-high conditions, had relatively lower ozone in DFW (45.3 ppbv) resulting from transport of lower background ozone from the Gulf of Mexico. For clusters that produce southeasterly or southwesterly winds over Houston, ozone values in DFW were always larger than those in Houston. Further, to determine the potential impact of Houston pollution on DFW ozone, a sensitivity simulation with no Houston emissions and a base simulation were performed. The difference between the simulations revealed ozone enhancements of 1–2 ppbv and coincident enhancements in NOy under south-southeasterly wind conditions. From these results, we conclude that downwind pollution from Houston and the Barnett Shale area exacerbates DFW ozone concentrations, underscoring the impacts of specific wind patterns on air quality in DFW.

scholarly journals SLIC Superpixel Based Self Organizing Maps Algorithm for Segmentation of Microarray Images

2018 ◽  
Vol 7 (1) ◽  
pp. 78
Author(s):  
Durga Prasad Kondisetty ◽  
Mohammed Ali Hussain
Keyword(s):  
Computational Cost ◽  
Microarray Technology ◽  
Qualitative And Quantitative Analysis ◽  
Clustering Methods ◽  
Self Organizing Maps ◽  
Microarray Image ◽  
Segmentation Quality ◽  
Som Algorithm ◽  
Simple Linear Iterative Clustering ◽  
Self Organizing

We can find the simultaneous monitoring of thousands of genes in parallel Microarray technology. As per these measurements, microarray technology have proven powerful in gene expression profiling for discovering new types of diseases and for predicting the type of a disease. Gridding, Intensity extraction, Enhancement and Segmentation are important steps in microarray image analysis. This paper gives simple linear iterative clustering (SLIC) based self organizing maps (SOM) algorithm for segmentation of microarray image. The clusters of pixels which share similar features are called Superpixels, thus they can be used as mid-level units to decrease the computational cost in many vision applications. The proposed algorithm utilizes superpixels as clustering objects instead of pixels. The qualitative and quantitative analysis shows that the proposed method produces better segmentation quality than k-means, fuzzy c-means and self organizing maps clustering methods.

scholarly journals Acoustic lung signals analysis based on Mel frequency cepstral coefficients and self-organizing maps

2016 ◽  
Vol 25 (43) ◽  
pp. 73-82
Author(s):  
Álvaro David Orjuela-Cañón ◽  
Hugo Fernando Posada-Quintero
Keyword(s):  
Neural Networks ◽  
Clustering Algorithm ◽  
Mel Frequency Cepstral Coefficients ◽  
Self Organizing Maps ◽  
Number Of Clusters ◽  
Supervised Training ◽  
Computer Aided ◽  
Unsupervised Training ◽  
Cepstral Coefficients ◽  
Self Organizing

This study analyzes acoustic lung signals with different abnormalities, using Mel Frequency Cepstral Coefficients (MFCC), Self-Organizing Maps (SOM), and K-means clustering algorithm. SOM models are known as artificial neural networks than can be trained in an unsupervised or supervised manner. Both approaches were used in this work to compare the utility of this tool in lung signals studies. Results showed that with a supervised training, the classification reached rates of 85 % in accuracy. Unsupervised training was used for clustering tasks, and three clusters was the most adequate number for both supervised and unsupervised training. In general, SOM models can be used in lung signals as a strategy to diagnose systems, finding number of clusters in data, and making classifications for computer-aided decision making systems.

scholarly journals ELECTROFACIES CLASSIFICATION OF PONTA GROSSA FORMATION BY MULTI-RESOLUTION GRAPH-BASED CLUSTERING (MRGC) AND SELF-ORGANIZING MAPS (SOM) METHODS

2020 ◽  
Vol 38 (1) ◽  
pp. 52
Author(s):  
Felipe Vasconcelos dos Passos ◽  
Marco Antonio Braga ◽  
Thiago Gonçalves Carelli ◽  
Josiane Branco Plantz
Keyword(s):  
Gamma Ray ◽  
Clustering Algorithms ◽  
Clustering Methods ◽  
Self Organizing Maps ◽  
Resolution Graph ◽  
Lithology Prediction ◽  
Paraná Basin ◽  
Parana Basin ◽  
Graph Based Clustering ◽  
Self Organizing

ABSTRACT. In Ponta Grossa Formation, devonian interval of Paraná Basin, Brazil, sampling restrictions are frequent, and lithological interpretations from gamma ray logs are common. However, no single log can discriminate lithology unambiguously. An alternative to reduce the uncertainty of these assessments is to perform multivariate analysis of well logs using data clustering methods. In this sense, this study aims to apply two different clustering algorithms, trained with gamma ray, sonic and resistivity logs. Five electrofacies were differentiated and validated by core data. It was found that one of the electrofacies identified by the model was not distinguished by macroscopic descriptions. However, the model developed is sufficiently accurate for lithological predictions.Keywords: geophysical well logging, lithology prediction, Paraná Basin. CLASSIFICAÇÃO DE ELETROFÁCIES DA FORMAÇÃO PONTA GROSSA UTILIZANDO OS MÉTODOS MULTI-RESOLUTION GRAPH-BASED CLUSTERING (MRGC) E SELF-ORGANIZING MAPS (SOM)RESUMO. Na Formação Ponta Grossa, intervalo devoniano da Bacia do Paraná, Brasil, restrições de amostragem são frequentes e interpretações litológicas dos registros de raios gama são comuns. No entanto, nenhum perfil geofísico único pode discriminar litologias sem ambiguidade. Uma alternativa para reduzir a incerteza dessas avaliações é executar uma análise multivariada combinando vários perfis geofísicos de poços por meio de métodos de agrupamento de dados. Nesse sentido, este estudo tem como objetivo aplicar dois algoritmos de agrupamento aos registros de raios gama, sônico e resistividade para fins de predição litológica. Cinco eletrofácies foram diferenciadas e validadas por dados de testemunhos. Verificou-se que uma classe identificada pelo modelo não foi identificada por descrições macroscópicas. Porém, o modelo é suficientemente preciso para predições litológicas.Palavras-chave: geofísica de poços, predição litológica, correlação rocha-perfil, Bacia do Paraná.

Sign in / Sign up

Export Citation Format

Share Document