SlideShare a Scribd company logo

Knime (Konstanz Information Miner)

•
•
K
Kiran Buriro

KNIME is an open source platform for data analytics and processing. It allows users to visually create data workflows using predefined nodes to perform tasks like data mining, analysis, transformation and more. KNIME can integrate data from multiple sources and interface with languages like R, Python and SQL. It provides a graphical user interface for building, executing and monitoring analytics workflows.

Related slideshows

Anomaly Detection and Spark Implementation - Meetup Presentation.pptx
Anomaly Detection and Spark Implementation - Meetup Presentation.pptxAnomaly Detection and Spark Implementation - Meetup Presentation.pptx
Anomaly Detection and Spark Implementation - Meetup Presentation.pptx
 
Knime
KnimeKnime
Knime
 
Zipline—Airbnb’s Declarative Feature Engineering Framework
Zipline—Airbnb’s Declarative Feature Engineering FrameworkZipline—Airbnb’s Declarative Feature Engineering Framework
Zipline—Airbnb’s Declarative Feature Engineering Framework
 
1 of 20
By: Kiran Buriro Assigned by: Sir Fida Chandio
What is KNIME ? • KNIME Stands for Konstanz Information Miner. • Developed at University of Konstanz in Germany 2004-2006 and focused initially on pharmaceutical research. • The KNIME is an open source platform for analytical data modelling and processing. • KNIME allows users to visually create data flows (or pipelines) • Written in Java based on the Eclipse SDK platform . • Modular platform for building and executing workflows using predefined components, called nodes. • Core functionality available for tasks such as standard data mining, analysis and manipulation. • GUI based with scripting integration. • An especially powerful aspect of KNIME is its ability to integrate data from multiple sources • KNIME also offers extensions that allow it to interface with R, Python, Java, and SQL.
KNIME DATA ANALYTICS LIFECYCLE READ DATA READ DATA READ DATA Extract, Transform, Load (ETL) Data Analytics or Predictive Analysis Reporting and/or Injection
KNIME GUI/WORK BENCH
KNIME GUI/WORK BENCH
A node is the smallest programming unit in KNIME Each node serves a dedicated task. After being created, a node needs settings to exec ute the task, this phase is called configuration. After configuration, a node needs to be executed to actually carry out the assigned task. 01 02 03 04 Node Status and Operations
Node Status and Operations • A node can have 3 states: Idle: The node is not yet configured and cannot be executed with its current settings. Configured: The node has been set up correctly, and may be executed at any time Executed: The node has been successfully executed. Results may be viewed and used in downstream nodes.
Node Status and Operations Input Output Status Partitioning Not Configured Idle Executed Error
Workflow
Workflow
Workflow
KNIME WORKFLOW • KNIME provides huge repository of modules for easy-to-use and for modular: KNIME Data Preprocessing Data fusion Data Transformation
DATABASE MySQL, any JDBC (Oracle, DB2, MySQL Server). FILES Csv, txt, Excel, Word, PDF, Images, texts. WEB,CLOUD Web services Twitter, Google FILESDATABASE WEB, CLOUD Data Access
KNIME ETL FEATURES ETL Logical joins Support for REGEX style replacements Rule-based filtering and transformation
Linear correlation and dependency measures Many nodes also support statistical standards such as count, sum, mean, etc. “Statistics” node has base measures of distribution KNIME STATISTICS
Data partitioning and multiple folds These are extended through partner implementations and scripting languages (R, Python, Weka, etc.) Base KNIME supports most machine learning algorithms KNIME MACHINE LEARNING
KNIME REPORTING • Generates reports in office document formats, PDF, and HTML • BIRT Tool as part of the Eclipse framework • Native part of the KNIME workbench • Extends data visualization capabilities • Auto-distribute by email, or publish to websites
 Process Mapping  Process Analysis IDEAS DATA AGGREGATION • Combine data from different sources, local or remote • ETL data into a single repository for querying/analytics BUSINESS INTELLIGENCE • Data intelligence and reporting over large aggregated datasets • Automated reusable workflows for standardized reporting PREDICTIVE ANALYTICS • Ability for insight across very large datasets KNIME ANALYTICS • Advantage of being a data agnostic aggregator • Ability to work through very large datasets with little hardware • Access to complex algorithms with easy tools DATA ANALYTICS USE CASES
KNIME ADVANTAGES • KNIMEs core-architecture allows processing of large data volumes that are only limited by the available hard disk space (not limited to the available RAM). E.g. KNIME allows analysis of 300 million customer addresses, 20 million cell images and 10 million molecular structures. • Additional plugins allows the integration of methods for Text mining, Image mining, as well as time series analysis. • KNIME integrates various other open-source projects, e.g. machine learning algorithms from Weka, the statistics package R project, ImageJ, and the Chemistry Development Kit . • KNIME is implemented in Java but also allows for wrappers calling other code in addition to providing nodes that allow to run Java, Python, Perl and other code fragments
Knime (Konstanz Information Miner)

More Related Content

Knime (Konstanz Information Miner)

  • 1. By: Kiran Buriro Assigned by: Sir Fida Chandio
  • 2. What is KNIME ? • KNIME Stands for Konstanz Information Miner. • Developed at University of Konstanz in Germany 2004-2006 and focused initially on pharmaceutical research. • The KNIME is an open source platform for analytical data modelling and processing. • KNIME allows users to visually create data flows (or pipelines) • Written in Java based on the Eclipse SDK platform . • Modular platform for building and executing workflows using predefined components, called nodes. • Core functionality available for tasks such as standard data mining, analysis and manipulation. • GUI based with scripting integration. • An especially powerful aspect of KNIME is its ability to integrate data from multiple sources • KNIME also offers extensions that allow it to interface with R, Python, Java, and SQL.
  • 3. KNIME DATA ANALYTICS LIFECYCLE READ DATA READ DATA READ DATA Extract, Transform, Load (ETL) Data Analytics or Predictive Analysis Reporting and/or Injection
  • 6. A node is the smallest programming unit in KNIME Each node serves a dedicated task. After being created, a node needs settings to exec ute the task, this phase is called configuration. After configuration, a node needs to be executed to actually carry out the assigned task. 01 02 03 04 Node Status and Operations
  • 7. Node Status and Operations • A node can have 3 states: Idle: The node is not yet configured and cannot be executed with its current settings. Configured: The node has been set up correctly, and may be executed at any time Executed: The node has been successfully executed. Results may be viewed and used in downstream nodes.
  • 8. Node Status and Operations Input Output Status Partitioning Not Configured Idle Executed Error
  • 12. KNIME WORKFLOW • KNIME provides huge repository of modules for easy-to-use and for modular: KNIME Data Preprocessing Data fusion Data Transformation
  • 13. DATABASE MySQL, any JDBC (Oracle, DB2, MySQL Server). FILES Csv, txt, Excel, Word, PDF, Images, texts. WEB,CLOUD Web services Twitter, Google FILESDATABASE WEB, CLOUD Data Access
  • 14. KNIME ETL FEATURES ETL Logical joins Support for REGEX style replacements Rule-based filtering and transformation
  • 15. Linear correlation and dependency measures Many nodes also support statistical standards such as count, sum, mean, etc. “Statistics” node has base measures of distribution KNIME STATISTICS
  • 16. Data partitioning and multiple folds These are extended through partner implementations and scripting languages (R, Python, Weka, etc.) Base KNIME supports most machine learning algorithms KNIME MACHINE LEARNING
  • 17. KNIME REPORTING • Generates reports in office document formats, PDF, and HTML • BIRT Tool as part of the Eclipse framework • Native part of the KNIME workbench • Extends data visualization capabilities • Auto-distribute by email, or publish to websites
  • 18.  Process Mapping  Process Analysis IDEAS DATA AGGREGATION • Combine data from different sources, local or remote • ETL data into a single repository for querying/analytics BUSINESS INTELLIGENCE • Data intelligence and reporting over large aggregated datasets • Automated reusable workflows for standardized reporting PREDICTIVE ANALYTICS • Ability for insight across very large datasets KNIME ANALYTICS • Advantage of being a data agnostic aggregator • Ability to work through very large datasets with little hardware • Access to complex algorithms with easy tools DATA ANALYTICS USE CASES
  • 19. KNIME ADVANTAGES • KNIMEs core-architecture allows processing of large data volumes that are only limited by the available hard disk space (not limited to the available RAM). E.g. KNIME allows analysis of 300 million customer addresses, 20 million cell images and 10 million molecular structures. • Additional plugins allows the integration of methods for Text mining, Image mining, as well as time series analysis. • KNIME integrates various other open-source projects, e.g. machine learning algorithms from Weka, the statistics package R project, ImageJ, and the Chemistry Development Kit . • KNIME is implemented in Java but also allows for wrappers calling other code in addition to providing nodes that allow to run Java, Python, Perl and other code fragments