BEGIN:VCALENDAR VERSION:2.0 PRODID:-//Drupal iCal API//EN X-WR-CALNAME:Events items teaser BEGIN:VEVENT UID:63f05a02d39a0 DTSTART;TZID=America/Toronto:20181031T130000 SEQUENCE:0 TRANSP:TRANSPARENT DTEND;TZID=America/Toronto:20181031T150000 SUMMARY:An introduction to feature selection CLASS:PUBLIC DESCRIPTION:Summary \n\nFeature selection is the process of selecting a sub set of relevant\nfeatures (commonly known as predictors or independent var iables) for\nmodel construction. Performing feature selection allows resea rchers to\nidentify irrelevant data\, improve the interpretation and incre ase\npredictive accuracy of learned models. A feature selection algorithm\ ncan be seen as the combination of a search technique for proposing new\nf eature subsets\, along with an evaluation which scores the different\nfeat ure subsets. The choice of evaluation measure heavily influences\nthe algo rithm. There are three main categories of feature selection\nalgorithms: w rappers\, filters and embedded methods. In this seminar\,\nwe will introdu ce some basic feature selection methods such as\nscore-based feature ranki ng\, stepwise subset selection and LASSO\nregression.\n\nRegistration\n[/s tatistical-consulting-and-collaborative-research-unit/node/45] is\nfree an d open to all University of Waterloo faculty\, staff\, graduate\nand under graduate students. The primary software we will discussed in\nthis seminar is RStudio. There is no hands-on work in this seminar.\n DTSTAMP:20230218T045426Z END:VEVENT END:VCALENDAR