Data mining problem types nach CRISP-DM

Quelle: CRISP-DM 1.0 Step-by-step data mining guide Pete Chapman (NCR), Julian Clinton (SPSS), Randy Kerber (NCR), Thomas Khabaza (SPSS), Thomas Reinartz (DaimlerChrysler), Colin Shearer (SPSS) and Rüdiger Wirth (DaimlerChrysler). © 2000 SPSS Inc. CRISPMWP-1104 CRISP-DM

problem type	appropriate technique
2.1 Data description and summarization	Iris: Attribute > pandas describe
2.2 Segmentation	Clustering techniques	Iris kNN 3D
	Neural networks
	Visualization
2.3 Concept descriptions	Rule induction methods
	Conceptual clustering
2.4 Classification	Discriminant analysis
	Rule induction methods	CRISP-DM, S.68 unten: If SEX = male and AGE > 51 then CUSTOMER = loyal ...
	Decision tree learning	iris decision tree
	Neural networks
	K nearest neighbor	https://medium.com/@srishtisawla/k-nearest-neighbors-f77f6ee6b7f5
	Case-based reasoning
	Genetic algorithms
2.5 Prediction	Regression analysis	analyticsvidhya ridge lasso > Regressionsgerade
	Regression trees
	Neural networks
	K nearest neighbor
	Box-Jenkins methods
	Genetic algorithms
2.6 Dependency analysis	Correlation analysis	https://en.wikipedia.org/wiki/Correlation_and_dependence > Beispiele
	Regression analysis
	Association rules	kdnugget > grocery transactions
	Bayesian networks	http://users.sussex.ac.uk/~christ/crs/kr-ist/lec09a.html > Reasoning as propagation
	Inductive logic programming
	Visualization techniques	Tableau: Market Basket Analysis / Heatmap

Entscheidungsbaum zur Auswahl von Algorithmen: scikit-learn.org > scikit-learn algorithm cheat sheet