KNIME INTRODUCTION

KNIME Introduction
 WHAT IS KNIME?

 KNIME stands for Konstanz Information Miner

 KNIME offers a complete platform for end-to-end data science, from creating analytic models, to
deploying them and sharing insights within the organization, through to data apps and services.
 It is an Open Source Tool used a GUI to assembly ‘nodes’ for data preprocessing (ETL), modeling, Data
Science, Machine Learning and Visualization
 Modules for:
 Data Mining
 Data Analysis
 Data Science
 Machine Learning
 More modules and extensions can be added!
 Written in Java and based on Eclipse
 WHERE TO GET IT AND OTHER ONLINE
RESOURCES

Download KNIME
[Link]
Skip the registration form, go straight to step (2) and download the version with
free extensions

Getting Started Guide | KNIME

[Link]
 WHAT CAN I DO WITH KNIME?

 Data Access
 File
 Database I/O

 Transformation
 Filtering, Grouping, Joining

 Analyze and Data mining

 Weka
 R /Python

 Data Science
 Machine Learning
 Visualization
 Different types of charts

 Deployment
 Text mining
 HOW DOES KNIME COMPARE WITH OTHERS?

Gartner’s Magic Quadrant for Advance Analytics Platforms

Leaders quadrant with SAS, IBM and Dell

Strong Performer / Contender in Forrester’s Wave

 KNIME LINGO

 Store your work in a workspace

 Workflows can contain nodes, meta nodes, connections, workflow variables, workflow credentials
and annotations
 Workspace can contain workflow groups built using the workflow editor
 Each node has a type, which identifies the algorithm associated with it
 Nodes have parameters, inports and
outports, and can have any of these states:
 Misconfigured
 Configured
 Queued for Execution
 Running
 Executed
 KNIME ANALYTICS PLATFORM

 Home
 Local Space
 KNIME Community Hub

 Workflow Tabs
 Workbench Editor
 Node Repository
 Workflow Space Explorer
 Workflow Monitor
 K-AI AI Assistant
GETTING SET UP WITH KNIME ANALYTICS PLATFORM
HOW TO BUILD A KNIME WORKFLOW

Search in Node Repository

Dragging nodes into Workflow Editor
Connecting Nodes
Configuring Nodes
Executing (per node or one-shot)
 INPUT AND OUTPUT PORTS

The input(s) is the data that the node

processes via the node ports, and the
output(s) are the resulting data.
Each node has specific settings, which you
can adjust in a configuration dialog.
Different types are represented by different
node ports.
Only ports of the same type indicated by
the same color can be connected.
 NODE STATUS

Node can be in four different states.

Node status is shown by a traffic light below each node.
WORKBENCH EDITOR WITH EXCEL READER
NODE
WORKFLOW WITH EXCEL SHEETS
PERFORMING K-MEANS CLUSTERING
DATA PREPROCESSING
EXAMPLE OF R SNIPPET
SIMPLE MODEL TRAINING FOR
CLASSIFICATION