Introduction to Neural Networks with R

Title: Introduction to Neural Networks with R
Speaker: Michael Grogan
Location: Room 304, Western Gateway Building, UCC

The presentation will be divided into three sections, covering the following areas:

• What is machine learning? In contrast with traditional statistical analysis, the focus of machine learning is to automate analysis. In other words, allow a computer to learn independently. What are the implications of this, and how is it impacting the area of data science as a whole?

• Supervised vs. Unsupervised learning. A machine learning model can learn in two fashions. Supervised - where a model is being used to discover a specific relationship from data. Unsupervised - an exploratory form of analysis where the relationship between data points is not yet known, and machine learning provides insights about data that had not previously been anticipated.

• Use of neural networks: Classification and regression. Neural networks are typically used to solve two main types of problems. Classification, where the model is looking to identify a specific category in the data. e.g. is this image an apple or a banana? On the other hand, regression problems are ones where a specific quantity is being calculated, e.g. how many cars is a car dealer forecasted to sell this year?

Biography

Michael Grogan is a machine learning consultant and educator who helps a variety of organisations solve business intelligence problems using Python and R.

He has presented his work at international conferences across Europe, and has published an R course for O'Reilly Media titled: Business Analytics with R — Statistics and Machine Learning. You can find out more about Michael's background and experience at www.michaeljgrogan.com.