Below is a description below of the different topics discussed during  the KNIME Courses.

 

Introductory Course

KNIME Overview

Once you have completed this session, you will be able to:

• install KNIME and its extensions;
• understand and use the KNIME workbench;
• find nodes and node information;
• create, configure, and execute basic workflows;
• inspect node results.

The concepts of node status, ports, and memory policy within KNIME are also described.

Visual Data Exploration

Here, you are introduced to the basics behind KNIME data visualization. This includes:

• interactive Views;
• Views Properties;
• the hiliting functionality (interactive brushing).

This session also gives some hints on how to perform string manipulation and type conversions in KNIME.

Training and Testing Models

The Training and Testing Models session provides you with the following knowledge.

• Data Preparation:
  • data partitioning into training and test set;
  • strategies to deal with missing values;
  • data normalization;
• Data Mining Models:
  • model training;
  • applying a trained model to new data;
  • a basic overview of the data mining techniques available in KNIME, like - but not limited to - Decision Trees, Clustering, Neural Networks, Naive Bayes Models, Linear Regression, etc ...
• Model Evaluation and Interpretation

Data Manipulation

This session focuses on the use of Data Manipulation nodes in KNIME, including nodes for:

• row- and column-wise transformations;
• row and column filtering;
• string manipulations;
• aggregations;
• convert and replace operations;
• date/time operations.

You will get an understanding of KNIME’s preprocessing nodes in order to modify and manipulate the data as needed.

Introductory Reporting

During this session you are given a short overview of the basic functionality of the report interface and you learn how to switch back and forth from the KNIME data analysis perspective to the KNIME reporting perspective.

In a step-by-step process, you are then shown how to build a small and easy report with basic components, like a table, a chart, a title, a logo, a page break, etc ... This session also includes the use of  maps and highlights.

Database Operations

This session introduces the whole topic of databases. In particular, it shows the nodes necessary to:

• connect to a database;
• build a SELECTquery;
• extract the data from the database;
• or alternatively write the data into the database.

Batch Execution

This session concludes the Introductory Course and explains how to run KNIME workflows in batch mode and how to configure KNIME to handle aggressive memory requests. 

 

 

Advanced Course

Flow Variables

This session teaches how to introduce external parameters into the workflow to better control it, by means of the flow variables. This means:

• a general description of what a flow variable is;
• the ways to introduce a new flow variable into the workflow;
• how to use a flow variable to control a node configuration;
• Quickform nodes.

Loops

Here the concept of loops is introduced. This session offers plenty of examples on how to repeat operation blocks inside a workflows. The following topics are discussed:

• The loop structure in a KNIME workflow;
• Loop specific commands;
• Description of basic loops;
• Alternative ways of collecting results;
• Looping on list of values.

Switches

This is the last session about flow control. Here we teach how to direct the data flow onto selected branches by using the Switch nodes, including the following topics:

• The concept of a switch block;
• The IF Switch blocks;
• The CASE Switch blocks.

Advanced Reporting

We explain how to recycle the flow variables into report parameters and how to build more professional and complex reports.

Integration of external Applications

KNIME offers the possibility to integrate external tools, like R, Weka, Java, Python, and even a customized external application of your choice. This session presents an overview of the main integration options in a KNIME workflow. This session has no hands-on exercises.

XML handling

This is the final session of the advanced course and explores how to extract information from XML structures.

 

 

Details for the KNIME Developer Training

The following topics are explained in detail during the KNIME developer training.

Architecture Overview

This session provides a general overview about the framework used to develop KNIME nodes within the integrated development environment (IDE) Eclipse.

Node Extension Wizard

This covers the KNIME Node Extension Wizard that generates a complete and ready-to-run node registered with a KNIME plug-in.

Data Types

This session describes the general data format used to transfer data between nodes. Each data table holds meta information about its structure in a so called data table spec. This structure contains the details (name, type, and domain info) for all columns within a table. KNIME supports a public API that allows the user-specific column data column type to be derived. This is explained based on examples during the session.

Dialog and Settings

This session covers the GUI components that can be used in KNIME to design customer dialogs in order to adjust node configuration. The underlying settings are explained that are transfered between node dialog and model.

Loop Support

This session describes how parts of a workflow can be executed multiple times. This feature of KNIME is known as loop support. We discuss the details of this functionality, covering flow and workflow variables in particular.

Node View

KNIME views are interactive in that they support hiliting capabilities (brushing) across different views. Within this session, a sample node view is implemented and registered, with the framework allowing interactions with all available views in KNIME.

Plug-in Deployment

This session shows how the developed KNIME node plug-in can be deployed from a KNIME development environment into a KNIME product.