Group 1 Data Access and Data Manipulation

Solution to the tasks for Group 1 in KNIME Data Science Learnathon
- Access data
- Preprocess data by filtering rows, filtering columns, converting column types, and handling missing values
- Join data from two different sources
- Generate new features by binning and by a rule
- Remove outliers
- Normalize data
- Partition data into a training and a test set
- Write data into a file

External resources

• Analytics - Model Selection to Predict Flight Departure Delays
• Will They Blend? The Blog Post Collection
• Missing Values
• Dimensionality Reduction and Feature Selection
• 7 Techniques for Data Dimensionality Reduction
• Four Techniques for Outlier Detection
• Four Techniques for Outlier Detection
• Normalization
• Partitioning
• KNIME E-Learning Course - Data Manipulation
• KNIME Analytics: a Review
• Outlier Detection in Medical Claims
• Original Airline Dataset