Publicação
Categories’ churn: a machine learning approach in retail
| Resumo: | In the realm of business, the cessation of relations with a company by its cus tomers or can yield profound financial repercussions. Such consequences manifest as diminished revenue and profitability, while also bringing risk to the company’s rep utation. Thus, comprehending the roots of customer churn is of major importance. Equally crucial is the formulation of effective strategies to mitigate churn, thereby enhancing customer satisfaction, retention, and overall profitability. Within the framework of this dissertation, clustering techniques were deployed alongside machine learning methodologies. This combination helped achieve a bal ance between accuracy and simplified model communication. In addition, data anal ysis techniques were deployed within the context of churn analysis, where total churn emerged as the optimal solution for the project, even if a case for partial churn could be made. The outcome was the development of a cohort of models capable of identifying the majority of churners within the selected pilot categories. Which empowers the retention department to implement more impactful retention campaigns, ensuring a more effective response to customer churn and bolstering overall business stability. |
|---|---|
| Autores principais: | Sousa, Carlos Filipe Fernandes |
| Assunto: | Churn Machine learning Clustering Classification Retail Big Data Abandono Aprendizagem automática Agrupamento Classificação Retalho |
| Ano: | 2023 |
| País: | Portugal |
| Tipo de documento: | dissertação de mestrado |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade do Minho |
| Idioma: | inglês |
| Origem: | RepositóriUM - Universidade do Minho |
| Resumo: | In the realm of business, the cessation of relations with a company by its cus tomers or can yield profound financial repercussions. Such consequences manifest as diminished revenue and profitability, while also bringing risk to the company’s rep utation. Thus, comprehending the roots of customer churn is of major importance. Equally crucial is the formulation of effective strategies to mitigate churn, thereby enhancing customer satisfaction, retention, and overall profitability. Within the framework of this dissertation, clustering techniques were deployed alongside machine learning methodologies. This combination helped achieve a bal ance between accuracy and simplified model communication. In addition, data anal ysis techniques were deployed within the context of churn analysis, where total churn emerged as the optimal solution for the project, even if a case for partial churn could be made. The outcome was the development of a cohort of models capable of identifying the majority of churners within the selected pilot categories. Which empowers the retention department to implement more impactful retention campaigns, ensuring a more effective response to customer churn and bolstering overall business stability. |
|---|