Learning what is important: feature selection and rule extraction in a virtual course
作者： Terence A. Etchells , Alfredo Vellido , Paulo J. G. Lisboa , Àngela Nebot , Francisco Mugica
DOI:
关键词: Machine learning 、 Extraction (chemistry) 、 Feature selection 、 Course (navigation) 、 Artificial intelligence 、 Pattern recognition 、 Computer science
摘要: Virtualcampusenvironmentsarebecomingamainstr eamalternativeto� traditionaldistancehighereducation.�TheInternet �mediumtheyuseallowsthe� gatheringofinformationonstudents'�usagebehavio ur.�Theknowledgeextracted� fromthisinformationcanbefedbacktothee/lear ningenvironmenttoease� teachers'�workload.�Inthiscontext,�twoproblemsa readdressedinthecurrent� study:�findingwhichusagefeaturesarebestatpre dictingonlinestudents'�marks,� andexplainingmarkpredictionintheformofsimpl eandinterpretablerules.�To� thateffect,�twomethodsareused:�FuzzyInductive� Reasoning�(FIR)�forfeature� selectionandOrthogonalSearch/BasedRuleExtraction � (OSRE).� Experiments� carriedoutontheavailabledataindicatethatstu dents'�markscanbeaccurately� predictedandthatasmallsubsetofvariablesexpl ainstheaccuracyofsuch� prediction,�whichcanbedescribedthroughasetof �actionablerules.��
researchgate.net 参考文章(4)
S. Ventura, C. Romero, Data Mining in E-learning ,(2006)
George J. Klir, Doug Elias, Architecture of Systems Problem Solving ,(2011)
Josep Maria Mirats i Tur, François E. Cellier, Rafael M. Huber, VARIABLE SELECTION PROCEDURES AND EFFICIENT SUBOPTIMAL MASK SEARCH ALGORITHMS IN FUZZY INDUCTIVE REASONING International Journal of General Systems. ,vol. 31, pp. 469- 498 ,(2002) , 10.1080/0308107021000042471
H. Tsukimoto, Extracting rules from trained neural networks IEEE Transactions on Neural Networks. ,vol. 11, pp. 377- 389 ,(2000) , 10.1109/72.839008