KNIME Analytics Platform

Creating Data Science

KNIME Analytics Platform is the open source software for creating data
science. Intuitive, open, and continuously integrating new developments,
KNIME makes understanding data and designing data science workflows and
reusable components accessible to everyone.

With nodes covering all stages of the data science life cycle, thousands of
ready to run example workflows, hundreds of components, a comprehensive
range of integrated tools, and the widest choice of advanced algorithms
available, KNIME Analytics Platform is the perfect platform for creating data
science. Our steady position on unrestricted open source provides access to a
global community of expertise and their contributions. KNIME Analytics Platform for Data Science Creation and KNIME Server for
Data Science Operations.

Build End to End Data Science Blend Data and Tools Leverage Machine Learning and AI

Build visual workflows Blend data Build machine learning models

Create workflows using the intuitive, drag Combine simple text files, databases, For classification, regression, dimension
and drop style graphical interface - documents, images, networks, even data reduction, or clustering, using advanced
without the need for coding. based on Apache Hadoop within the same algorithms including deep learning, tree-
visual workflow. based methods, and logistic regression.
Blend data from any source
Open and combine simple text formats Integrate R and Python Portable and durable
(CSV, PDF, XLS, JSON, XML, etc.), Include R and Python code in your KNIME Optimize model performance with
unstructured data types (images, workflows, reusing expertise that is hyperparameter optimization, boosting,
documents, networks, molecules, etc.), or graphically documented and shared bagging, stacking, or building complex
time series data. among data scientists. ensembles.

Validate models
Choose from nodes and components Blend different tools
Apply performance metrics including
Build your workflow, model each step of Blend state of the art tools easily: Integrate
Accuracy, R², AUC, and ROC. Perform cross
your analysis, control the flow of data, and Apache Hadoop, Spark, and MLlib using
validation to guarantee model stability.
ensure your work is always current. the KNIME Big Data Integrations.
Additional integrations include deep
Make predictions
Use pre-made blueprints learning frameworks and other machine
Use validated models directly, or with
Select one of the thousands of publicly learning libraries (H2O, Weka, and more).
industry leading PMML, including Apache
available workflows from the KNIME Hub,
Spark.
or use the integrated workflow coach to Document visually
guide you through building your own Coding is optional and work is
workflow. documented visually with the easy to learn
graphical interface.

Nodes for the Entire Data Science Life Cycle

• Connectors for all major file formats and

databases

• Native and in-database data blending

and transformation

• Support for a wealth of data types such

as XML, JSON, images, documents,
networks, time series, and more

• Advanced predictive and machine

learning algorithms

• Integrations with state of the art

machine learning libraries: H2O, Keras
for Deep Learning, Scikit-Learn, and
others

• Interactive data views and reporting KNIME Analytics Platform covers all stages of the data science life cycle.
using web-based methods
 Connect Seamlessly from KNIME Analytics Platform with Extensions and Integrations

A large variety of KNIME Extensions provide access to complex data types, plus advanced machine learning algorithms. KNIME
Analytics Platform and KNIME Server integrate seamlessly with many other open source projects.

Access Apache Hadoop data as well as Hadoop data storage such as Hive and Impala. Model and run Apache Spark jobs and access
the power of scalable analytics in local KNIME environments. Use pieces of R or Python code within a KNIME workflow, build a
predictive model, apply that model to new data, or simply create other types of visualizations. Read, create, edit, train, and execute
deep neural networks. Keras provides access to deep learning frameworks such as TensorFlow, CNTK, and others. Integrate data from
Amazon Redshift, H2, Hive, Impala, Microsoft SQL, MySQL, Oracle, PostgreSQL, and more.

Community Extensions are open source contributions from other KNIME users and are accessible from within KNIME Analytics
Platform. Partner Extensions offer a commercial set of capabilities, ranging from industry specific applications to sophisticated,
scientific software integrations - all created and maintained by KNIME Trusted Partners. Users can package their own code into
KNIME by using the KNIME SDK environment and also contribute to the KNIME Community.

External tools and software that integrate seamlessly with KNIME Analytics Platform.

KNIME Server: Productionizing Data Science

KNIME Server complements KNIME Analytics Platform and is the enterprise software for team based collaboration, automation,
management, and deployment of data science workflows as analytical applications and services. KNIME Server comes in three editions
for organizations and teams of all sizes and enables the following enterprise functionality:

• Integrated Deployment for closing the gap between creation and production of data science without manual intervention
• Automatic workflow execution with the ability to view, edit, and execute workflows remotely
• Guided Analytics for building applications and giving users the appropriate level of interaction at any stage of the life cycle
• Governance and compliance through the explainability / interpretability of models, GDPR compliance, metadata mapping, and more
• Flexible cloud and hybrid deployments to meet computing and budgeting needs - also with auto-scaling functionality

At KNIME, we build software for fast, easy, and

intuitive access to advanced data science, helping
organizations drive innovation. For over a decade, a
thriving community of data scientists in over 60
countries has been working with our platform on
every kind of data: from numbers to images,
molecules to humans, signals to complex networks,
and simple statistics to big data analytics. Our
headquarters are based in Zurich, with additional
offices in Konstanz, Berlin, and Austin. We’re open
for innovation®, so visit us at KNIME.com.

KNIME Software: one ecosystem.

KNIME AG
Hardturmstrasse 66 info@knime.com
8005 Zurich, Switzerland www.knime.com

© 2020 KNIME AG. All rights reserved.

The KNIME® trademark and logo and OPEN FOR INNOVATION® trademark are used by KNIME AG under license from KNIME GmbH, and are registered in the United States and the European Union.
All third party trademarks (including logos and icons) referenced, remain the property of their respective owners.