Mineração de dados e análise do comportamento do consumidor: estudo de caso em um website de ensino a distância

• Ano de defesa: 2010
• Autor(a) principal: Kippes, Alexander
• Orientador(a): Zambalde, André Luiz
• Banca de defesa: Antonialli, Luiz Marcelo, Souki, Gustavo Quiroga
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Lavras
• Programa de Pós-Graduação: DAE - Programa de Pós-graduação; Pós-graduação stricto-sensu em Administração
• Departamento: Departamento de Administração e Economia
• País: BRASIL
• Palavras-chave em Português: Mineração de dados; Comportamento do consumidor; Marketing; Data Mining; Web user Mining; Consumer behavior; Web Mining usuário
• Área do conhecimento CNPq: CNPQ_NÃO_INFORMADO
• Link de acesso: https://repositorio.ufla.br/handle/1/1992

The main and general objective of this research was investigating the Web Usage Mining process of the Latu senso on distance in Software Engineering Free (ESL), Federal University of Lavras - MG (UFLA), aimed at generating information and knowledge for research, development and decision-directed the actions, practices and marketing activities and customer relationships in organizations providing educational services in the distance mode. The methodology for this research is technological nature, qualitative and quantitative approaches, with goals of exploratory and descriptive, using a case study procedures and operations research (mathematical analysis). In addition the interviews with the responsible person of the marketing in the organization accompanied by bibliographic and documentary were collected and extracted data related to the visitor behavior on every page of the website architecture components ESL using specific software tools. There were collected and extracted related data to visitor browsing behavior (user) on every page of the website architecture components "ESL" in two specific time periods. The Web mining approach allowed the extraction of data from real user browsing, for any click during the entry, exit and navigation of the website. The data passed through the intermediate processes of cleaning, preparation and storage, for later extraction of browsing patterns, mathematical and statistical analysis and generation of information and so knowledge. After the first Period, was preliminarily proposed the restructuring of the website, and again proceeded a new strategy of data mining and visitors behavior discovery patterns in order to verify the impact of restructuring carried out in accordance return to the organization.The results indicated that the data mining process of the user behavior on the Web is complex and the organization was not properly prepared. The challenges and pitfalls encountered drew attention to important general considerations applied to organizations that work in distance education. The main aspect to consider is the proper planning of the relationship between conventional marketing, Web marketing and staff responsible for information technology. The search for interaction between teams and relationship marketing and those who are responsible for Web mining technology in organizations is crucial for the preparation of plans and activities and e-marketing, and consequently to the expansion of business. Anyway, the main bottlenecks to be addressed are considering the context of Web usage mining and marketing are: (1) the used key word in search engines, (2) the placement of sequential pages of the website aimed to create a funnel that takes the visitor on a registration form or registration, (3) the architecture and design appropriate and attractive website, (4) the interaction between offline marketing and online marketing, for example, Google AdWords, and finally, (5) the appropriate choice of reference pages such as social media and other Web pages. Finally, it should be noted that practically all these bottlenecks are present in the context of the case study.