Análisis de clúster para segmentar en el mercado de una empresa utilizando una red neuronal artificial
Cargando...
Archivos
Fecha
2014
Autores
Profesor/a Guía
Facultad/escuela
Idioma
es
Título de la revista
ISSN de la revista
Título del volumen
Editor
Universidad Andrés Bello
Nombre de Curso
Licencia CC
Licencia CC
Resumen
En la actualidad las organizaciones operan en mercados muy competitivos donde el proceso de segmentar, clasificar o dividir su mercado, se vuelve indispensable para poder comprender mejor las necesidades de sus consumidores y así satisfacerlas de mejor forma. Además de permitirles crear ventajas frente a sus competidores y facilitar el ajuste de oferta en productos y servicios.
Es entonces que, basado en la necesidad de las empresas por clasificar su mercado, que el siguiente trabajo realiza un análisis de clúster para segmentar el mercado de una empresa utilizando las redes neuronales de Kohonen, es decir, las redes SOM (Self-Organizing Map).
Los resultados del análisis de clúster permitieron segmentar el mercado de una empresa fabricante de acero en Chile, definiendo 3 segmentos (Clústeres) para un mercado total de 89 empresas constructoras consumidoras de acero.
A partir de los resultados obtenidos también se logró definir un proceso para la segmentación de mercado utilizando una red SOM y se pudo concluir que para obtener una compresión total de los efectos de la clusterización realizada, se debe tener un conocimiento del mercado del acero y la construcción, para poder entender el porqué de las asociaciones dentro de las empresas de cada segmento.
In the present-day organizations operate in highly competitive markets where the process to segment, classify or divide their market, becomes essential for better understanding of the needs of their consumers and thus satisfy them best. Besides affording them advantages over their competitors and to facilitate adjustment of the offer of products and services. It is then that, based on the need for companies to classify their market, the following paper does a cluster analysis to segment the market of a company using neural networks of Kohonen, i.e., the SOM networks (Self -Organizing Map) . The results of cluster analysis allowed to segment the market of a steel manufacturing company in Chile, defining 3 segments (clusters) for a total of 89 construction companies consumers of the steel market. From the results obtained it was also possible to define a process for market segmentation using a SOM network and it was concluded that to obtain a complete understanding of the effects of clustering performed, should have a knowledge of the steel market and construction in order to understand why partnerships within the companies in each segment.
In the present-day organizations operate in highly competitive markets where the process to segment, classify or divide their market, becomes essential for better understanding of the needs of their consumers and thus satisfy them best. Besides affording them advantages over their competitors and to facilitate adjustment of the offer of products and services. It is then that, based on the need for companies to classify their market, the following paper does a cluster analysis to segment the market of a company using neural networks of Kohonen, i.e., the SOM networks (Self -Organizing Map) . The results of cluster analysis allowed to segment the market of a steel manufacturing company in Chile, defining 3 segments (clusters) for a total of 89 construction companies consumers of the steel market. From the results obtained it was also possible to define a process for market segmentation using a SOM network and it was concluded that to obtain a complete understanding of the effects of clustering performed, should have a knowledge of the steel market and construction in order to understand why partnerships within the companies in each segment.
Notas
Tesis Ingeniería en Automatización y Robótica
Palabras clave
Segmentación del Mercado, Redes Neurales (Ciencia de la Computación)., Análisis de Clúster, Kohonen, Mapa Auto-Organizativo