Lecture 1: Data Mining Concepts

Question 1

What is data mining?

DM is technique used to __ valuable information from __ __ by discovering __, __ and __.

Answer 1

DM is technique used to extract valuable information from large dataset by discovering patterns, relationships and anamolies.

Question 2

KDD

Knowledge Discovery in Databases (KDD) process consists of:

1. Data C__ & I__
2. Data S__ & T__
3. Data M__
4. E__ & P__
5. Knowledge

Answer 2

1. Data Cleaning & Integration
2. Data Selection & Transformation
3. Data Mining
4. Evaluation & Presentation
5. Knowledge

Question 3

What is the concept of DM functionalities?

Function that __ and __ classes for __ predictions.

Answer 3

Function that describes and distinguish classes for future predictions.

Question 4

Association Analysis

What is confidence and support?

Confidence: The percentage that ___ __ occurs ___ of __.
Support: The __ percentage that ____ & __ occus __.

Answer 4

Confidence: The percentage that event B occurs because of event A.
Support: The total percentage that event A & B occus together.

Question 5

Discovering ‘interesting’ patterns.

Objective vs Subjective measures:

Objective: Based on s__ and s__ of p__.

Subjective: Based on u__ b__ in data.

Answer 5

Objective: Based on statistics and structures of patterns. E.g. support and confidence for association rules.
Subjective: Based on user’s belief in data. E.g. unexpectedness, novelty.