Preprocessing features of TADA-Ed

1
Preprocessing features of TADA-Ed

• Agathe Merceron
• Kalina Yacef

2
Login Qid Mistake Concept
DavidM 924 X A
DavidM 172 X C
DavidM 172 Y A
DavidM 172 Z D
MaryT 926 Y D
MaryT 154 X B
JohnS 924 Z B
JohnS 924 X A
JohnS 924 Y A
JohnS 924 X A
JohnS 165 Z C
JohnS 154 W D
First we choose an attribute to aggregate the
data. Lets suppose we want to compare students
Login ??
DavidM
MaryT
JohnS
3
Login Qid Mistake Concept
DavidM 924 X A
DavidM 172 X C
DavidM 172 Y A
DavidM 172 Z D
MaryT 926 Y D
MaryT 154 X B
JohnS 924 Z B
JohnS 924 X A
JohnS 924 Y A
JohnS 924 X A
JohnS 165 Z C
JohnS 154 W D

• Chosen attribute to aggregate login
• Preprocessing performed
• Qid Number of types

Login Qid
DavidM
MaryT
JohnS
4
Login Qid Mistake Concept
DavidM 924 X A
DavidM 172 X C
DavidM 172 Y A
DavidM 172 Z D
MaryT 926 Y D
MaryT 154 X B
JohnS 924 Z B
JohnS 924 X A
JohnS 924 Y A
JohnS 924 X A
JohnS 165 Z C
JohnS 154 W D
Now we want to only consider the total number of
mistakes each student made

• Chosen attribute to aggregate login
• Preprocessing performed
• Qid Number of types
• Mistake Total Number

Login Qid mistake
DavidM
MaryT
JohnS
5
Login Qid Mistake Concept
DavidM 924 X A
DavidM 172 X C
DavidM 172 Y A
DavidM 172 Z D
MaryT 926 Y D
MaryT 154 X B
JohnS 924 Z B
JohnS 924 X A
JohnS 924 Y A
JohnS 924 X A
JohnS 165 Z C
JohnS 154 W D
We want to know with what concepts the student
made mistakes

• Chosen attribute to aggregate login
• Preprocessing performed
• Qid Number of types
• Mistake Total Number
• Rule Nominal

Login Qid mistake A B C D
DavidM
MaryT
JohnS