A carregar...
Publicação
NOS & Nova SBE Project-Based Learning - predicting the volume of customers to flag for call center’s specialized team : enhancing the performance of time series models through unsupervised outlier detection and treatment techniques
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 2.18 MB | Adobe PDF |
Autores
Orientador(es)
Resumo(s)
This thesis is part of a broader research effort extending the outcomes of the year-long
Project-Based Learning initiative with NOS, a prominent telecommunications company in
Portugal, focusing on optimizing the number of clients that should be flagged for specialized
call center teams, to increase clients’ satisfaction. While the larger thesis addresses
both model performance and explainability, this specific work focuses on improving model
performance through outlier detection. By refining the handling of outliers, this study contributes
to more accurate predictions, ultimately enhancing the effectiveness of client selection.
Descrição
Palavras-chave
Call centers Time series forecasting Prediction modeling Unsupervised outlier detection Trimming Winsorization Robust estimation SARIMA XGBoost