Recommender System for Predicting Student Performance:
We compare recommender system techniques with
traditional regression methods such as logistic/linear regression by using educational data for intelligent tutoring
systems. Experimental results show that the proposed approach can improve prediction results.
Supervisor : Mohammad Fadi Taqi Al-Din.
https://www.shamra.sy/academia/show/5b0b46c82b551
10. المشكلة صياغة:
•let S be a set of student
• T be a set of task
• f ⊆ R be a performance
measure
• then D ⊆ (S × T × f) is the
triple data collected from the
computer-aided tutoring
systems
•Given s ∈ S and t ∈ T, our
problem is to predict f
11. Data Sets and Method
تماستخدام:
KnowledgeDiscoveryandDataMining(KDD)Challenge20101
12. على التدريب بيانات تحتوي كما
التالية: features
• Incorrects
• corrects
• Hints
• time information
• Target : correct first attempt (CFA)
13. التوصية نظام مع الطالب أداء بيانات مطابقة:
تمتمقابلة
user => student
rating => correct first attempt (CFA)
: itemsتمتبالمسألة مقابلته)التمرين(خي عدت لدينا يوجد حيثارات
من(data set) KDDمن بعدد تمثيلها يمكن والتيcombination :
PH : problem hierarchy
PN : problem name
SN : step name
PV : problem view