Chapter 20

Question 1

What are types of data duplication on non-unique primary key

Answer 1

• Duplicate identification numbers
• House holding
• Individualization

Question 2

What are 2 sub types of Duplicate identification numbers

Answer 2

• Multiple customer numbers

- Multiple employee numbers

Question 3

What is House holding: data duplication problem

Answer 3

Multiple people living in same one house. (e.g. there are many people and families living in one home. All have bank accounts. When bank send advertise to its account holders, it will be sending multiple brochures to one house)

Question 4

What is individualization: data duplication problem

Answer 4

There is one person behind but looks more than one. (e.g. Mr ahad, Major. ahad)

Question 5

What is purge

Answer 5

Remove duplicate records

Question 6

What 0 and 1 represents in degree of similarity

Answer 6

1 - near

0 - far

Question 7

What is BSN method

Answer 7

Basic sorted neighborhood method is a method for removal of duplication record. It has 3 steps.
1- Identify attributes of record that will make a key of each record.
2- Sort that record on the basis of key.
3- Data is organized in a set. Each set is called window. Each window representing number can help us to identify duplicates.

Question 8

What is BSN window

Answer 8

Data is organized in a set. Each set is called window . Each window representing number can help us to identify duplicates. BSN window size minimum will be 2 and maximum can be entire list.

Question 9

What is BSN method: selection of keys

Answer 9

In this method, we select which parts of attributes of records and those part will help to make key. That key will be sorted and do BSN.

Question 10

What is BSN method: matching candidates

Answer 10

Merging of duplicate records is complex inferential process