With growing amounts of data we need new methods for data analysis. This three-day-class gives an overview over the most important datamining-methods. Unsupervised learning methods try to find structures in your data without an explicit response variable. Typical unsupervised methods are principal component- and cluster-analysis. The main topic in this class are prediction models. This includes logistic regression, decision trees and neural networks. The class concludes with the general idea of ensemble-prediction-models.
- Uses cases in data mining
- Clustering (KMEANS, Hierarchical clustering)
- Principal component analysis (PCA)
- Prediction modeling (logistic regression, decision trees, neural networks)
- Concept of ensemble-models
Duration of the training: 3 days