Social networks group properties

somdn_product_page

(Downloads - 0)

Catégorie :

For more info about our services contact : help@bestpfe.com

Table of contents

Chapter 1 Introduction 
1.1 Research context
1.2 Structure of social networks
1.3 Research on social network privacy analysis
1.4 Motivations and challenges of this work
1.5 Contributions of the thesis
1.6 Outline
Chapter 2 Defining sensitive subjects
2.1 Introduction
2.2 Conducting a survey on the behaviour of French users of social media
2.3 Analysing responses and defining sensitive subjects
2.4 Possible attack vectors according to the behaviour of participants
2.5 Conclusions
Chapter 3 Disclosing friendship and group membership links
3.1 Introduction
3.2 Modelling social network for on-line link disclosure attacks
3.3 Problematics and objectives
3.4 Social networks group properties
3.5 Link disclosure attacks
3.6 Conclusions
Chapter 4 Overview of our implemented prediction system
4.1 Introduction
4.2 Architecture
4.3 SONSAI user guide
4.4 Examples of inference scenarios
4.5 Conclusions
Chapter 5 Sampling and modelling social networks
5.1 Introduction
5.2 Definitions
5.3 Sampling social network around a target user profile
5.4 Modelling discovered links and nodes by graphs
5.5 Anonymizing the social network graph models
5.6 Conclusions
Chapter 6 Cleansing the collected data
6.1 Introduction
6.2 Definitions
6.3 Motivations for cleansing data
6.4 Cleansing the sensitive graph
6.5 Cleansing the learning graphs
6.6 Cleansing results
6.7 Conclusions
Chapter 7 Analysing cleansed data and inferring target sensitive values
7.1 Introduction
7.2 Translating social attributed network to a text document
7.3 Applying Word2Vec to compute node embeddings
7.4 Inferring hidden sensitive values of the target user profile
7.5 Measuring inference accuracy
7.6 Experiments and results
7.7 Conclusions
Chapter 8 Conclusions and perspectives
Appendixs 
Appendix A Datasets 
A.1 Dataset 1 (D1)
A.2 Dataset 2 (D2)
Appendix B Questionnaire 
Bibliography

Laisser un commentaire

Votre adresse e-mail ne sera pas publiée. Les champs obligatoires sont indiqués avec *