Master's degree thesis

Thesis title: "Identifying the source of false information in social networks".

Related publication: "Contrasting the Spread of Misinformation in Online Social Networks"

Project description

• EdmondsMemoryOpt.py - Customised version of the Edmonds' algorithm with less memory/time usage compared to networkx
• EdmondsLowMemory.py - Version that uses the disk space to store intermediate results
• imeterOpt.py - Imeter-Sort algorithm included in the publication "Sources of misinformation in online social networks: Who to suspect?", Nam P. Nguyen Dung T. Nguyen and My T. Thai. S
• independent_cascade_opt.py - An implementation of the independent cascade model
• camerini.py - Implementation of the algoritm included in the publication "The k best spanning arborescences of a network", L. Fratta P. M. Camerini and F. Maffioli (1980)
• test.py - Test class, compares Imeter-Sort with the Edmonds' algorithm accuracy under the assumption that there is only one source of false information
• test_multi_sources.py - Test class, compares Imeter-Sort with Camerini's algorithm (with an euristic applied which deals also with not connected graphs) whereas there are multiple source of false information

Addition resources:

• thesis.pdf
• Presentation.pdf