Intelligent Decision Support Systems
RapidMiner
Curriculum Development
of Master’s Degree Program in
Industrial Engineering for Thailand Sustainable Smart Industry
Module 2: Software Tools for IDSS Development
• The analytic hierarchy process (AHP)
• R-software
• RapidMiner
• WEKA
• Deep Learning for Smart Production
RapidMiner
Why RapidMiner
• RapidMiner’s data science platform delivers lightning fast business
impact for over 40,000+ organizations in every industry to drive
revenue, reduce costs, and avoid risks.
https://rapidminer.com/why-rapidminer/
Why RapidMiner
Old Way RapidMiner Way
•Slow to predict opportunities & •Lightning fast business impact
risk •Depth for data scientists,
•Highly specialized expertise simplified for everyone else
•Full transparency & governance
•Opaque modeling for machine learning
•Legacy fragmented tools •End-to-end collaboration
•Vendor lock-in platform
•Open source & extensible
RapidMiner
RapidMiner
RapidMiner
RapidMiner
RapidMiner
RapidMiner Studio:
Visual workflow designer for the entire analytics team
Visual Workflow Designer
•Increase productivity across the entire data science team, from analysts to
experts
•Speed up and automate the creation of predictive models in a drag + drop
visual interface
•Rich library of 1500+ algorithms and functions ensures the best model for
any use case
•Pre-built templates for common use cases including customer churn,
predictive maintenance, fraud detection, and many more
•“Wisdom of Crowds” provides proactive recommendations at every step to
help beginners
Connect to Any Data Source
• Work with all of your data, no matter where it lives
• Instantly create point + click connections to databases, enterprise
data warehouses, data lakes, cloud storages, business applications
and social media
• Easily re-use connections any time and easily share them with
anyone who needs access
• Connect to new sources with extensions from the
RapidMiner Marketplace
Automated In-database Processing
• Run data prep and ETL inside databases to keep your data
optimized for advanced analytics
• Query and retrieve data without writing complex SQL
• Harness the power of highly scalable database clusters
• Supports MySQL, PostgreSQL, and Google Big Query
Data Visualization & Exploration
• Evaluate data health, completeness, and quality
• Understand patterns, trends, and distributions with scatter plots,
histograms, line charts, parallel coordinates, box plots, and more
• Quickly find and fix common data quality problems including missing
values and outliers
• Explore data using robust statistical overviews and over 30
interactive visualizations
Data Prep & Blending
• Eliminate the hassle of preparing data for predictive modeling
• RapidMiner Turbo Prep offers a fully interactive point + click data
prep experience
• Extract, join, filter, and group data across any number of sources
• Create repeatable data prep and ETL processes that can be
scheduled and shared
Visual & Automated Machine Learning
• Quickly create impactful machine learning models without writing code
• RapidMiner Auto Model creates models in 5 clicks using automated
machine learning
• Choose from hundreds of supervised and unsupervised machine learning
algorithms
• Implement basic and advanced ML techniques including regression,
clustering, time-series, text analytics, and deep learning
• Build the model to be sensitive to constraints like costs to optimize the
predictions for your desired business impact
• Use both automated and manual feature engineering to optimize model
accuracy
Model Validation
• Understand the true performance of a model before deploying to
production
• Eliminate overfitting through a unique approach that prevents model
training pre-processing data from leaking into the application of the
model
• Add proven techniques, like cross validation, to a model with just a
single mouse click
Explainable Models Not Black Boxes
• Create visual data science workflows that are easy to explain and
easy to understand
• Each step in the data prep, modeling, and validation process is
documented for complete transparency
• Visuals are easy to explain to others in the organization to gain buy-
in for deployment
• Supports the Local Interpretable Model-Agnostic Explanations
(LIME) framework
Get More From R & Python Code
• Scalable code deployment and collaboration between coders and non-
coders
• Deploy code-based models and code-containing models into a scalable
platform
• Eliminate duplicate work and ensure others re-use your work by
uploading your code snippets into the RapidMiner repository for use in a
simplified visual workflow designer
• Make use of the vibrant Python ecosystem by building on top of
specialized and cutting-edge libraries and making it available for the rest
of your team using RapidMiner
• Leverage data prep and ETL pipelines others have produced
Flexible Scoring & Model Operations
• Turn predictive insights into business impact
• Quickly deploy scored data to spreadsheets and data visualization
tools or turn models into production web services with RapidMiner
Server
• RapidMiner Model Ops delivers an easy way for less advanced
users to put models into production and properly manage them
• Integrate with enterprise scheduling tools
• Add RapidMiner Real-Time Scoring for demanding high transaction /
low latency use cases
Automation & Process Control
• Build sophisticated visual workflows and automate important tasks
• Use process control operators to create workflows that repeat and
loop over tasks, branch flows and access system resources
• Supports a variety of scripting languages for custom integrations
and automatons
• Schedule processes
Open & Extensible
• Integrate with existing applications and code
• Use R and Python code and libraries to extend RapidMiner
• Integrate RapidMiner with Python through the
RapidMiner Python library on Github
• Download new functionality provided by RapidMiner, our partners,
and community through the RapidMiner Marketplace
• Add new capabilities to RapidMiner through its flexible extension
mechanism
• RapidMiner Studio open core is available under the AGPL license
https://rapidminer.com/products/studio/
RapidMiner: By Industry
• Automotive • Insurance
• CPG, Retail & e-Commerce • Life Sciences
• Communications • Manufacturing
• Energy • Travel, Transport & Logistics
• Financial Services • Utilities
• Healthcare
RapidMiner: By Use Case
• Churn Prevention • Predictive Maintenance
• Customer LTV • Product Propensity
• Customer Segmentation • Quality Assurance
• Demand Forecasting • Risk Management
• Fraud Detection • Text Mining
• Next Best Action • Up- and Cross-Selling
• Price Optimization
RapidMiner: Case Studies
RapidMiner: Case Studies
RapidMiner
RapidMiner
download on https://rapidminer.com/get-started/
RapidMiner
RapidMiner
คูม ื การใช ้ Rapodminer
่ อ
การสอนใช ้ Rapidminer
• https://www.youtube.com/watch?v=XJ-qHAjRRy0
Rapid Miner IDSS Research
