Machine Learning
Date:
25/01/2017
Organised by:
London School of Economics and Political Science
Presenter:
Professor Kenneth Benoit
Level:
Intermediate (some prior knowledge)
Contact:
Esti Sidley, 0207 955 6947, methodology.admin@lse.ac.uk
Map:
View in Google Maps (WC2A 2AE)
Venue:
PhD Academy, 4th floor, Lionel Robbins Building, Portugal Street, London
Description:
This workshop will introduce students to the fundamentals of machine learning, beginning with a contrast between stastistical model fitting in a classical sense, versus predictive models using "statistical learning" methods typically designed to classify outcomes. We will cover the foundations of machine learning concepts, such as how to partition data into training and test sets; how to evaluate model performance, including precision, recall, and accuracy; how to cross-validate models; and how to assess model fit. Models surveyed will supervise machine learning, such as regression and logistic regression, k-nearest neighbour, and Naive Bayes; and unsupervised methods such as clustering models, decision trees, and principal components analysis. Because this course is brief, we will survey these methods rather than cover them in great depth, but the objective is to provide an overview to the field of machine learning that students will be able to build on if they wish to continue with a particular aspect. All methods will be demonstrated with working code in the R language using social, political, or economic data examples.
Cost:
£30
Website and registration:
http://eshop.lse.ac.uk/product-catalogue/methodology/events/machine-learning-workshop
Region:
Greater London
Keywords:
R, Machine learning, , regression, , logistic regression, , k-nearest neighbour, , Naive bayes
Related publications and presentations: