Hybrid Clustering Approach
Hassan discusses combining LSTM autoencoder and dynamic time warping in a hybrid approach for clustering customer data, emphasizing the importance of selecting proper features over the clustering method for meaningful results.In this clip
From this podcast
Data Skeptic
Customer Clustering
Related Questions
What would be the right approach to go about feature selection and correlation analysis for predicting the number of units sold or the % growth in units sold over the next 3 years in a time series problem?
How can clustering methods be extended in the context of the episode Causal Inference in Educational Systems and the clip Time Series Clustering?