Introduction to Predictive Analytics
Dean Abbott, Co-Founder and Chief Data Scientist, SmarterHQ
Predictive Analytics and the related fields of Data Science and Machine Learning have become an integral part of organizations who want to become “data-driven”. But there is more to predictive analytics that data and algorithms. This workshop is aimed at business professionals and managers who are used to looking at data and trying to make sense of data but want to move beyond reporting and summary statistics (like averages) to gain deeper insight into their data.
The workshop will be lecture + demo format. For demos, I will be using the open source software KNIME (http://www.knime.org), a visual programming tool for Advanced Analytics. All workflows used in demonstrations will be made available to attendees before the workshop. Any attendees who would like to gain practical insights into predictive analytics are encouraged to install the software before the workshop and download the workflows (at a URL to be provided before the workshop).
Attendees will learn:
• What predictive analytics is, how those within the field define it, and how it differs from other data-centric disciplines such as statistics and business intelligence
• A project-oriented framework for predictive analytics in six steps
• Special considerations for setting up data
• The top algorithms used in building predictive models: regression, decision trees, and neural networks
• How to assess predictive model accuracy
• How to interpret predictive models