Visualization of high dimensional data set
Post on 13-Apr-2017
450 Views
Preview:
Transcript
Visualization of High Dimensional Data Set
SRGE Workshop, Cairo University Conference Hall (19-September-2015)
Hanaa Ismail Elshazly
http://www.egyptscience.net
Overview
Introduction Problem Definition Motivation
Related Work Proposed Approaches Results and Discussion Conclusion and Future Works
SRGE Workshop, Cairo University Conference Hall (19-September-2015)
2
Introduction
3
SRGE Workshop, Cairo University Conference Hall (19-September-2015)
Problem Definition
Decision rules suffer of some shortcomings like Difficulty of knowledge acquisition process. Big size of rules that need maintenance. More cost development and time from experts and knowledge
engineers.
The need for an automatic system for good decision rules selected will offer great help and will reduce cost and effort for knowledge base construction and rules maintenance.
SRGE Workshop, Cairo University Conference Hall (19-September-2015)
4
Motivation
Decision rules present an easy and strong method of inference consistent with expert knowledge and ability of expression and explanation.
Decision rules can be easily adapted due to its declarative representation.
SRGE Workshop, Cairo University Conference Hall (19-September-2015)
5
Related Work
6
SRGE Workshop, Cairo University Conference Hall (19-September-2015)
Related Work [Couturier et al, 2007] presented a graphical-based visualization prototype
(CBVAR) to exploit generic bases which can reduce the number of rules. The generic bases include the most relevant rules containing minimal premises and maximal conclusions.
[Wong et al, 1999] presented a novel association-rule visualization system to tackle the text mining and visualization research. The goal is studying the relationships and implications among topics in a corpus and discover association rules that relate between some topics. The system includes two text engines to ex tract conceptual topics , the first is word-based and the second is concept-based. The visualization technique uses the 2D matrix to depict the rule-to-item relationship.
[Buono and Costabile, 2005] present a visual strategy that exploits a graph-based technique and parallel coordinates to visualize the results of association rule mining algorithms. The paper illustrates a framework for Visual Data Mining called DAE (Data Analysis Engine) as a result of the work within the European funded project FairsNet
SRGE Workshop, Cairo University Conference Hall (19-September-2015)
7
Proposed Approach
8
SRGE Workshop, Cairo University Conference Hall (19-September-2015)
Proposed Approach
SRGE workshop in Cairo University Conference Hall (12-September-2015)
9+
Results and Discussion
10
SRGE Workshop, Cairo University Conference Hall (19-September-2015)
Results and Discussion
SRGE Workshop, Cairo University Conference Hall (19-September-2015)
11
Results and Discussion
SRGE Workshop, Cairo University Conference Hall (19-September-2015)
12
Results and Discussion
SRGE workshop in Cairo University Conference Hall (12-September-2015)
13
Visualization of 1000 rules
Results and Discussion
SRGE workshop in Cairo University Conference Hall (12-September-2015)
14
Visualization of 500 rules
Results and Discussion
SRGE workshop in Cairo University Conference Hall (12-September-2015)
15
Visualization of 98 rules
Results and Discussion
SRGE workshop in Cairo University Conference Hall (12-September-2015)
16
Conclusion and Future Works
17
SRGE Workshop, Cairo University Conference Hall (19-September-2015)
Conclusion and Future Works
In this study, a hybrid method for breast cancer diagnosis based on GA and RS is presented.
The hybrid proposed method exhibits consistent and better performance than the RS classifier only.
The hybrid proposed method provides the expert by a promising tool to extract knowledge visually and instantly.
SRGE Workshop, Cairo University Conference Hall (19-September-2015)
18
Thanks and Acknowledgement19
SRGE Workshop, Cairo University Conference Hall (19-September-2015)
top related