1. SylabUZ
2. Faculty of Mathematics, Computer Science and Econometrics
3. summer term 2023/2024
4. Data Engineering - Second-cycle studies leading to MS degree
5. Practical Applications of Data Mining Systems

Generate PDF for this page

Practical Applications of Data Mining Systems - course description

Aim of the course

• Acquire the modeling skills required to understand and store big data in big data sets.
• Using skills to make decisions such as cancer detection, fraud detection, customer segmentation and machine downtime prediction.
• Learning about the data mining process and modeling techniques using one IBM SPSS Modeler program.
• Creating models based on selected data, testing models with historical data, using current data.

Prerequisites

Save changes

Fundamental of statistics.

Scope

1. Introduction to data mining
   • CRISP-DM methodology
   • Introduction to SPSS Modeler - a predictive data mining workshop
   • SPSS Modeler interface
2. Data retrieval process
   1. Understanding the business
   2. Understanding data
   3. Data preparation
3. Modeling techniques
   1. Introduction to modeling techniques
   2. Cluster analysis (unsupervised learning)
   3. Classification and prediction (supervised learning)
   4. Classification, training and testing
   5. Sampling in classification
   6. Predictive Modeling Algorithms in SPSS Modeler
   7. Automatic selection of algorithms
4. Model evaluation
   1. Performance evaluation data
   2. Accuracy as a performance evaluation tool
   3. Overcoming accuracy limits
   4. ROC Curves
5. Implementation on IBM Bluemix
   1. Evaluating new data
   2. Implementation of a predictive model
   3. What is IBM Bluemix?
   4. Predictive Modeling: Cloud Deployment
   5. SPSS Collaboration and Implementation Services

Teaching methods

Conventional lecture, problem-based lecture. Laboratory exercises. Discussion.

Learning outcomes and methods of theirs verification

Outcome description	Outcome symbols	Methods of verification	The class form

Assignment conditions

The grade for the laboratory will be based on the results from the colloquium and/or projects (80%) and activity in classes (20%).

Recommended reading

1. Axel Buecker, Theresa Morelli, Colin Shearer IBM SPSS predictive analytics: Optimizing decisions at the point of impact An IBM Redguide publication 2010

Further reading

Notes

Modified by dr Maciej Niedziela (last modification: 11-04-2024 16:06)