Charles VERNEREY Thesis defense

Address:

Campus de Nantes - Amphi Georges Besse

Mr Charles Vernerey from the Automation, Production and Computer Sciences Department (DAPI) and LS2N laboratory, will present his research about :

"Multi-objective modeling and optimization for knowledge discovery"

 

Thesis defense notice

See more

The objective of this thesis is to develop innovative approaches to effectively represent user preferences in multi-objective decision contexts for the extraction of interesting knowledge.
Firstly, we introduce a new Constraint Programming (CP) model to efficiently extract Pareto optimal patterns (also known as skypatterns) that scales effectively.
Next, we demonstrate how skypatterns can be leveraged to extract high-quality, non-redundant association rules without the need to set thresholds, in contrast to existing state-of-the-art approaches.
We then go beyond Pareto optimality to represent user preferences by introducing a novel approach based on the Choquet integral, a complex aggregation function that accounts for interactions (positive or negative) between the measures used to evaluate pattern quality. Lastly, we present a new Java-based CP library for modeling and solving several pattern mining problems.

Organizer(s)

Thesis acreditation from IMT Atlantique with the Doctoral School SPIN

 

Keywords : Constraint Programming, Multi-Objective Optimisation, Pattern Mining, Preference Learning

Published on 26.10.2023
 
1
 
2
 
3
 
4
 
5
 
6
 
7
 
8
 
9
 
10
 
11
 
12
 
13
 
14
 
15
 
16
 
17
 
18
 
19
 
20
 
21
 
22
 
23
 
24
 
25
 
26
 
27
 
28
 
29
 
30
 
31