A carregar...
Publicação
Segmentação de Produtos usando técnicas de Clustering Estudo de Caso numa loja de museu
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 12.19 MB | Adobe PDF |
Autores
Orientador(es)
Resumo(s)
"Em Marketing, segmentação é uma parte essencial em qualquer estratégia operacional de
Marketing, (...), Segmentação tem vindo a tornar-se mais complexa na era da Indústria 4.0
e virá a implicar desperdícios de recursos se não aplicada corretamente. Em termos de tec-
nologia, existem vários novos métodos para implementação de segmentação, recorrendo
a diversos algoritmos de Machine Learning (ML)". Atualmente, com as necessidades dos
consumidores demasiado variadas, e uma disponibilidade de dados, como nunca antes
vista, capaz de tornar qualquer análise num problema de altas dimensões, é necessário
uma abordagem diferente, que envolva técnicas analíticas e computacionais. A presente
dissertação, tem como objectivo, investigar as vantagens para as organizações na apli-
cação de segmentação, as variáveis usualmente usadas; quais as vantagens das técnicas
de ML para a segmentação, face às técnicas tradicionais, as metodologias usualmente
usadas e como os resultados da segmentação podem ser utilizados para gerar valor nas
organizações.
Para atingir o objetivo proposto, foi realizado um levantamento da literatura sobre o
tema, em específico, investigando a temática de segmentação, Marketing Mix e o uso de
ML em Marketing Research, de forma a encontrar resposta às questões enunciadas acima.
De forma a comprovar as respostas obtidas através da literatura foram desenvolvidos
vários modelos de análise de Clusters e foi realizado um caso de estudo em parceria com
Fundação Calouste Gulbenkian (FCG).
Com a presente dissertação foi permitido concluir que as técnicas de ML são ferra-
mentas com potencial no que toca a analisar elevadas quantidades de dados. De entre as
hipóteses existentes, constatou-se que técnica mais usada para segmentação é a análise
de Clusters; com a segmentação de produtos é possível, através do Marketing Mix, definir
estratégias operacionais para as organizações alocarem eficientemente os seus recursos.
"In marketing, segmentation is an essential part in any business marketing strategy, (...), Segmentation becomes more complex in the 4.0 industry era and it will generate waste of resources if not well implemented. In terms of technology, there are many new im- plementations of segmentation with several ML algorithms". Nowadays, with consumer needs so varied, and an unprecedented availability of data that makes any analysis a high dimensional problem, a different approach is needed, involving analytical and computa- tional techniques. This dissertation aims to investigate the advantages for organizations in the application of segmentation, the commonly used variables; what are the advantages of ML techniques for segmentation, compared to traditional techniques, the methodolo- gies usually used and how the results of segmentation can be used to generate value in organizations. To achieve the proposed goals, a survey of the literature on the subject was carried out, specifically, investigating the topic of segmentation, Marketing Mix and the use of ML in Marketing Research, to find an answer to the questions set out above. In order to prove the answers obtained through the literature, several models of Clusters analysis were developed and a case study was carried out in partnership with FCG. With this dissertation it was possible to conclude that ML techniques are tools with potential when it comes to analyzing large amounts of data. Among the existing models, it was found that the most used technique for segmentation is the Clusters analysis; with the segmentation of products it is possible, through the Marketing Mix, to define operational strategies for organizations in order to allocate their resources.
"In marketing, segmentation is an essential part in any business marketing strategy, (...), Segmentation becomes more complex in the 4.0 industry era and it will generate waste of resources if not well implemented. In terms of technology, there are many new im- plementations of segmentation with several ML algorithms". Nowadays, with consumer needs so varied, and an unprecedented availability of data that makes any analysis a high dimensional problem, a different approach is needed, involving analytical and computa- tional techniques. This dissertation aims to investigate the advantages for organizations in the application of segmentation, the commonly used variables; what are the advantages of ML techniques for segmentation, compared to traditional techniques, the methodolo- gies usually used and how the results of segmentation can be used to generate value in organizations. To achieve the proposed goals, a survey of the literature on the subject was carried out, specifically, investigating the topic of segmentation, Marketing Mix and the use of ML in Marketing Research, to find an answer to the questions set out above. In order to prove the answers obtained through the literature, several models of Clusters analysis were developed and a case study was carried out in partnership with FCG. With this dissertation it was possible to conclude that ML techniques are tools with potential when it comes to analyzing large amounts of data. Among the existing models, it was found that the most used technique for segmentation is the Clusters analysis; with the segmentation of products it is possible, through the Marketing Mix, to define operational strategies for organizations in order to allocate their resources.
Descrição
Palavras-chave
Segmentação de Produtos Loja de Retalho Loja de Museu Clustering Machine Learning Marketing