Data Science and Predictive AI Fundamentals.(using Modelling Software)

This course introduces you to the principles and practice of data science, including using AI algorithms for predictive analytics,  which are illustrated using the CRISP-DM methodology. The course provides training in the basics of how to import, explore, and prepare data with IBM SPSS Modeler and introduces the learner to modelling, and the fundamentals of what different modeling techniques are used and when without having to build or programme yourself.  We introduce the learners to the importance of programming languages such as R and Python in Data Science but it is not a course focused on programming languages and building your own algorithms but rather on understanding fundamentals, what the different algorithms are for, applying best practice methodologies and a chance to "hands-on" use powerful software with embedded and verified algorithms for Data Science projects. This foundation course is all about understanding data and best practices in  data science and being good analyst/ researcher. It will be a good basis then to deciding what further studies/ career pathway to take and we will advise what further studies. Non-Programmers and programmers alike can become good and principled data scientists.

We use IBM® SPSS® Modeler to demonstrate Data Science as it is a set of data mining tools that enable you to quickly develop predictive models using business expertise/academic knowledge and deploy them to improve decision making. Designed around the industry-standard CRISP-DM model, IBM SPSS Modeler supports the entire data mining process, from data to better business results.

IBM SPSS Modeler offers a variety of modeling methods taken from machine learning, artificial intelligence (AI), and statistics. The methods available on the Modeling palette allow you to derive new information from your data and to develop predictive models. Each method has certain strengths and is best suited for particular types of problems.

Who is it for?

This course is aimed for business analysts, data scientists, researchers, students and individuals who are at the Foundation level in Data Science and  IBM SPSS Modeler or want to find out more about using such tools and techniques.   

What should I have?

An understanding of your business/academic research data is recommended. Good overall IT competency and a good basic knowledge of Excel would be helpful. An understanding of the fundamentals of statistical and data analysis would be helpful but not essential and the trainer will adapt to the prior learning level of the participants. 

Our trainer is an experienced IBM-SPSS Statistics and DataMining Consultant

Refreshments and course notes are included. Delivery is online/in-person and/or hybrid depending on the preferences of each group and the facilities available.

Course Content

1. Introduction to data science

• List two applications of data science
• Explain the stages in the CRISP-DM methodology
• Describe the skills needed for data science

2. Introduction to IBM SPSS Modeler (as an example of a data science tool)

• Describe IBM SPSS Modeler's user-interface
• Work with nodes and streams
• Generate nodes from output
• Use SuperNodes
• Execute streams
• Open and save streams
• Use Help

3. Introduction to data science using IBM SPSS Modeler

• Explain the basic framework of a data-science project
• Build a model
• Deploy a model

4. Collecting initial data

• Explain the concepts "data structure", "of analysis", "field storage" and "field measurement level"
• Import Microsoft Excel files
• Import IBM SPSS Statistics files
• Import text files
• Import from databases
• Export data to various formats

5. Understanding the data

• Audit the data
• Check for invalid values
• Take action for invalid values
• Define blanks

6. Setting the of analysis

• Remove duplicate records
• Aggregate records
• Expand a categorical field into a series of flag fields
• Transpose data

7. Integrating data

• Append records from multiple datasets
• Merge fields from multiple datasets
• Sample records

8. Deriving and reclassifying fields

• Use the Control Language for Expression Manipulation (CLEM)
• Derive new fields
• Reclassify field values

9. Identifying relationships

• Examine the relationship between two categorical fields
• Examine the relationship between a categorical field and a continuous field
• Examine the relationship between two continuous fields

10. Introduction to modeling

• List three types of models
• Use a supervised model
• Use a segmentation model

*Disclaimer: Kindly note the scheduled dates below are tentative and are therefore subject to change. Please, do register your interest as we are taking provisional bookings.