Training M1

• RapidMiner and RapidAnalytics Overview
• Data Import
• Repository

The training starts with an introduction to the RapidMiner's Workbench and
shows how to efficiently utilize the graphical user interface to manage
projects, processes, data sets and results. During the lesson you will be guided
through the different available perspectives and views. After learning the
basics, it is shown how to go the first step towards all Data Mining Tasks:
Import the data into the integrated repositories.

These integrated repositories are then described in more detail and it is
shown how to configure RapidAnalytics Repositories for coordinating efficient
team work and deployment of results.

Training M2

• Data Mining with RapidMiner
• Cross Validation
• Attribute Selection

After we got comfortable with the RapidMiner and RapidAnalytics Workbench, this lesson
introduces into data mining. Together with some background knowledge from
statistical learning theory first Data Mining processes are designed. For
assessing how well these process fit the problem, the Cross-Validation is
introduced as most common and reliable statistical tool for performance estimation.

With the gained knowledge we can start to get first insight into our data:
Which are the most important attributes? Does sex or age determine if one is
interested in soccer?

All this will be imparted by having hands-on exercises for designing processes which
illustrate the theoretical background.

Training M3

• Data Preprocessing
• Process Design and Meta Data Propagation
• Quick Fixes

After the last lesson thought you how to perform Data Mining, we have to
admit, that most data in real world isn't in the condition to be directly used
for Data Mining. This lesson now will demonstrate the various preprocessing
capabilities to prepare your data for the final mining. During this you will
learn to appreciate the new Meta Data Propagation, which will help you a lot
during process design. It will keep you informed about the effects of your
operators, offers QuickFixes to repair broken parts of the process. All of this
is done in background and in real time, without actually processing the possible
millions of records in your data.

Training T1

Time Series Analysis and Forecasting

• Data Preprocessing for Forecasting Scenarios
  
• Numerical Predictions
  
• Validation of Forecasting
  

This lesson will tackle the following type of tasks: A time series, i.e. a
collection of measurements for different time points is given, and one searches
the most probable value of this measurement for future time points. This
includes the demonstration of necessary preprocessing steps like numerical data
transformations, an introduction how to apply previously introduced statistical
regression methods, neural networks, or support vector machines (SVM), and a
discussion of validation methods in order to estimate the goodness of the
predictions. These methods are especially useful for numerical predictions from
series data as they often occur in financial markets but also in production
settings and many others.

Training T2

Text Mining with RapidMiner

• Preprocessing of Unstructured Data
  
• Text Analysis
  
• Applications of Text and Web Mining
  

This lesson will shift the focus to knowledge discovery from unstructured
data like text documents. It will show the necessary preprocessing steps and the
most successful methods for automatic text classification (including Naive Bayes
and Support Vector Machines, SVM) and text clustering. Practical exercises for
different settings that frequently occur in different settings like for example
e-mail spam detection, automatical e-mail routing, adaptive personal news
filtering, sentiment analysis of text documents like news, web pages, blogs,
e-mail or PDF documents will enable the participants to transfer the previously
gained knowledge of general data mining to text mining problems.

Training T3

Market Basket Analysis

• Data Preprocessing for Transaction Data
  
• Finding Itemsets and Association Rules
  
• Visualization and Application of Models
  

During this lesson it is demonstrated how to analyse data from customer
transactions to better understand the needs of your customers, to predict their
behaviour, and to increase sales and improve profitability by leveraging cross-
and up-selling opportunities. Together with the necessary preprocessing steps, the
fast and highly effective operators of RapidMiner for shopping basket analysis
are introduced. After the patterns have been extracted, the practical steps
which are necessary to create and exploit such models are exercised.