Disambiguating novice object oriented programming errors using similarity measures
Intelligent Tutoring Systems (ITSs) are computer programs for educational support that can diagnose problems of individual learners (Sison, 2000). They make use of student models in order to tailor interactions to individual learners. There have been many approaches taken to solve this problem. One...
Saved in:
| Main Author: | |
|---|---|
| Format: | text |
| Language: | English |
| Published: |
Animo Repository
2007
|
| Subjects: | |
| Online Access: | https://animorepository.dlsu.edu.ph/etd_masteral/3571 https://animorepository.dlsu.edu.ph/context/etd_masteral/article/10409/viewcontent/CDTG004372_P.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | De La Salle University |
| Language: | English |
| Summary: | Intelligent Tutoring Systems (ITSs) are computer programs for educational support that can diagnose problems of individual learners (Sison, 2000). They make use of student models in order to tailor interactions to individual learners. There have been many approaches taken to solve this problem. One such solution is the use of a bug library to model the set of student errors. Since it is difficult to identify all student errors and to manually build the bug library, dynamic bug libraries have been used. Dynamic bug libraries can automatically build its bug library during run time by accepting student errors inputted into the system and adding it into the library. A key element in clustering is the similarity measure. The similarity measure identifies to which set a new object will be clustered into the bug library. Works like MEDD (Sison, 2000), which is a student modeling system, focuses on logic programming in Prolog and shows the effectiveness of using such an approach. Javabugs (Cruz, 2003) is a student modeling system that uses the same approach as MEDD that is applied to object oriented programming in Java. The results of Javabugs show that the technique it applies can not disambiguate constantly co-occurring yet unrelated errors. This research aims to use a different similarity measure that will allow the disambiguation of errors in object oriented programs more accurately. Keywords: Student Modeling, Bug Library, Clustering, Similarity measure. |
|---|