Powered by OpenAIRE graph
Found an issue? Give us feedback
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/ Repositório do ISCTE...arrow_drop_down
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
versions View all 1 versions
addClaim

Árvores de decisão e redes neuronais: Aplicação a web mining

Authors: Henriques, Marcos André Pais;

Árvores de decisão e redes neuronais: Aplicação a web mining

Abstract

A evolução do conceito de Marketing coloca a relação com o cliente no centro da estratégia da empresa. O fácil acesso a informação torna os clientes mais exigentes e susceptíveis à mudança de marcas com que se relacionam. Assim, as empresas sentem a necessidade de implementar estratégias de CRM – Customer Relationship Managemen que permitam obter informação e enviar estímulos aos clientes em todos os pontos de contacto destes com a empresa. Este trabalho explora o potencial da Internet enquanto ferramenta que permite obter conhecimento sobre os consumidores, centrando-se na análise de dados obtidos através do site de um Clube de Fidelização de uma marca de Grande Consumo. Assim, propõe-se com este trabalho uma metodologia de Web Mining de utilização que permita predizer comportamentos futuros, através de dados previamente recolhidos acerca dos comportamentos de utilização por parte dos utilizadores registados. Para tal, a metodologia proposta assenta em duas etapas: segmentação e modelação. Na segmentação dos utilizadores recorre-se ao algoritmo Two-Step, reflectindo os comportamentos ao longo de três anos após a data de registo. Para a modelação, recorre-se a Árvores de Decisão (algoritmo CART) e Redes Neuronais (algoritmo Backpropagation), como métodos de classificação. Propõe-se ainda, para além da utilização de cada método de forma individual, a combinação de ambos num Modelo Híbrido. Espera-se com esta metodologia obter informação que possibilite a incorporação em estratégia de CRM, nomeadamente, possibilitando criar políticas de Marketing geradoras de motivos de interesse e capazes de captar o retorno dos utilizadores ao Site de forma continuada.

The evolution of the Marketing concept puts the relationship with customer in a central position in the company strategy. Easy access to information makes customers more demanding and likely to change brands to which they relate. As an immediate consequence, the companies feel the need to implement new strategies for CRM - Customer Relationship Management in order to obtain information and send stimuli to customers at all points of contact with the company. This work explores the potential of the Internet as a tool to obtain knowledge about consumers, focusing on the analysis of data obtained through the site of a Loyalty Club for an FMCG brand. In order to achieve the main purpose, this work proposes to use a methodology of Web Usage Mining that allows to predict future behaviors, by considering data previously collected about the behavior of Registered Users. The implemented methodology is based on two steps: segmentation and modeling. In the first step the users are segmented by using the Two-Step algorithm, reflecting the behavior along three years after the date of registration. For modeling, we use the Decision Trees (CART algorithm) and Artificial Neural Networks (Backpropagation algorithm) as methods of classification. It is also proposed, besides the use of each method individually, the combination of both in a Hybrid Model. With this methodology, we expect to obtain information that facilitates the incorporation into CRM strategy, including the creation of Marketing policies that generate interest and are capable of capturing continuously the return of the Users Site.

C44; C45; M31

Country
Portugal
Keywords

Neural Networks, M31, M Business administration and business economics - Marketing - Accounting - Personnel economics, Decision Trees, Redes Neuronais, Árvore de decisão -- Decision tree, Redes neuronais -- Neural networks, Domínio/Área Científica::Engenharia e Tecnologia::Outras Engenharias e Tecnologias, Data mining --, CRM Customer Relationship Management --, C Mathematical and quantitative methods, CRM, Domínio/Área Científica::Ciências Sociais::Economia e Gestão, Data Mining, Árvores de decisão, C44, C45

  • BIP!
    Impact byBIP!
    selected citations
    These citations are derived from selected sources.
    This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    0
    popularity
    This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
    Average
    influence
    This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    Average
    impulse
    This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
    Average
    OpenAIRE UsageCounts
    Usage byUsageCounts
    visibility views 46
    download downloads 69
  • 46
    views
    69
    downloads
    Powered byOpenAIRE UsageCounts
Powered by OpenAIRE graph
Found an issue? Give us feedback
visibility
download
selected citations
These citations are derived from selected sources.
This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Citations provided by BIP!
popularity
This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
BIP!Popularity provided by BIP!
influence
This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Influence provided by BIP!
impulse
This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
BIP!Impulse provided by BIP!
views
OpenAIRE UsageCountsViews provided by UsageCounts
downloads
OpenAIRE UsageCountsDownloads provided by UsageCounts
0
Average
Average
Average
46
69
Green