BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Date iCal//NONSGML kigkonsult.se iCalcreator 2.20.4//
METHOD:PUBLISH
X-WR-CALNAME;VALUE=TEXT:ÌÇÐÄÔ­´´
BEGIN:VTIMEZONE
TZID:America/New_York
BEGIN:STANDARD
DTSTART:20191103T020000
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20200308T020000
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:calendar.379141.field_event_date.0@www.wright.edu
DTSTAMP:20260220T001002Z
CREATED:20191206T215601Z
DESCRIPTION:Ph.D. Committee:  Drs. Guozhu Dong (advisor)\, Keke Chen\, Kris
 hnaprasad Thirunarayan\, Pratik Parikh (BIE)\, and Hemant Purohit (George 
 Mason University) ABSTRACT:Many classification algorithms have been propos
 ed over the last 50 years.  However\, each classification algorithm may pe
 rform poorly for certain applications (or data sets)\, and it is of intere
 st to understand why and how such poor performance happens.  Poor performa
 nce can occur due to the data being especially hard to classify\, or becau
 se of inherent weaknesses in a type of classifier.In this dissertation we 
 answer these questions by examining the richness and simplicity of easily 
 correctable opportunities that are present when given classification algor
 ithms are applied.  We accomplish these by using a novel\, pattern-based a
 pproach to systematically identify weaknesses of different classifiers\, a
 nd to evaluate the potential of improvement on classification performance.
  Moreover\, we develop a number of useful metrics to consistently measure 
 and compare the richness and correctability of multiple significant opport
 unities.  It is hoped that the insights and techniques developed here can 
 provide guidance to future classification researchers\, and to machine lea
 rning practitioners concerning classifier weaknesses\, and can help them t
 o better understand what classifiers are best suited to given difficult-to
 -classify applications.
DTSTART;TZID=America/New_York:20191213T130000
DTEND;TZID=America/New_York:20191213T150000
LAST-MODIFIED:20191209T133126Z
LOCATION:304 Russ Engineering
SUMMARY:Ph.D. Dissertation Proposal Defense Pattern-based Analysis of Class
 ifier Weaknesses By Nicholas Skapura
URL;TYPE=URI:/events/phd-dissertation-proposal-defens
 e-pattern-based-analysis-classifier-weaknesses-nicholas
END:VEVENT
END:VCALENDAR