Icon

04. Data Mining

There has been no title set for this workflow's metadata.

There has been no description set for this workflow's metadata.

Activity II: Logistic RegressionTrain a model that predicts whether a wine is red or wine based on some chemical features.- Open the output table of the normalizer node to get familiar with the dataset.- Partition the wine data into training and test set (80% for training with Stratified Sampling on the column color).- Train a logistic regression model on the training set with Target Column = color.- Apply the model to the test set.- Evaluate the quality of the model with the Scorer (JavaScript) node or the Scorer node. Activity I: Decision Trees- Read the table Customer.table saved in the workflow group data in the KNIME Explorer.- Partition the customer data into training and test set (80% for training with Stratified Sampling on the columnTarget).- Train a decision tree on the training set with Class Column = Target and ReturnLastPrediction for the no true childstrategy.(Hint: Remember, you can change the no true child strategy in the second tab PMML Settings.)- Apply the model to the test set.- Evaluate the quality of the model with the Scorer (JavaScript) node or the Scorer node.- Try out different settings, e.g. with and without reduced error pruning, and see whether you can improve the modelaccuracy. z-normalize the data with0 mean and standarddeviation of 1Read the wine.tabledatasetCustomer.tableRelative 80%/20%Stratified SAmpling - TargetBest configurationso far - Min numbersrecords per node - 3PredictorOverall Acuracy - 86.42%Kappa - 0.728LearnerPredictorRelative 80%/20%Stratified SAmpling - ColorOverall Acuracy - 99.38%Kappa - 0.983 Normalizer (PMML) Table Reader(deprecated) Table Reader Partitioning DecisionTree Learner Decision TreePredictor Scorer (JavaScript) LogisticRegression Learner Logistic RegressionPredictor Partitioning Scorer (JavaScript) Activity II: Logistic RegressionTrain a model that predicts whether a wine is red or wine based on some chemical features.- Open the output table of the normalizer node to get familiar with the dataset.- Partition the wine data into training and test set (80% for training with Stratified Sampling on the column color).- Train a logistic regression model on the training set with Target Column = color.- Apply the model to the test set.- Evaluate the quality of the model with the Scorer (JavaScript) node or the Scorer node. Activity I: Decision Trees- Read the table Customer.table saved in the workflow group data in the KNIME Explorer.- Partition the customer data into training and test set (80% for training with Stratified Sampling on the columnTarget).- Train a decision tree on the training set with Class Column = Target and ReturnLastPrediction for the no true childstrategy.(Hint: Remember, you can change the no true child strategy in the second tab PMML Settings.)- Apply the model to the test set.- Evaluate the quality of the model with the Scorer (JavaScript) node or the Scorer node.- Try out different settings, e.g. with and without reduced error pruning, and see whether you can improve the modelaccuracy. z-normalize the data with0 mean and standarddeviation of 1Read the wine.tabledatasetCustomer.tableRelative 80%/20%Stratified SAmpling - TargetBest configurationso far - Min numbersrecords per node - 3PredictorOverall Acuracy - 86.42%Kappa - 0.728LearnerPredictorRelative 80%/20%Stratified SAmpling - ColorOverall Acuracy - 99.38%Kappa - 0.983Normalizer (PMML) Table Reader(deprecated) Table Reader Partitioning DecisionTree Learner Decision TreePredictor Scorer (JavaScript) LogisticRegression Learner Logistic RegressionPredictor Partitioning Scorer (JavaScript)

Nodes

Extensions

Links

Download

To use this workflow in KNIME, download it from the below URL and open it in KNIME:

Download Workflow
On NodePit since: 2023-09-30
Last update: 2024-11-21
Created with KNIME version: v5.1.1

Deploy, schedule, execute, and monitor your KNIME workflows locally, in the cloud or on-premises – with our brand new NodePit Runner.

Try NodePit Runner!