Predicting interactions between individuals with structural and dynamical informationArticleAuteurs : Thibaud Arnoux
1; Lionel Tabourier
1; Matthieu Latapy
1
NULL##NULL##0000-0002-0975-6109
Thibaud Arnoux;Lionel Tabourier;Matthieu Latapy
Capturing both structural and temporal features of interactions is crucial in many real-world situations like studies of contact between individuals. Using the link stream formalism to model data, we address here the activity prediction problem: we predict the number of links that will occur during a given time period between each pair of nodes. To do this, we take benefit from the temporal and structural information captured by link streams. We design and implement a modular supervised learning method to make prediction, and we study the key elements influencing its performances. We then introduce classes of node pairs, which improves prediction quality and increases diversity.
Volume : Vol. 5 - Analyse de graphes et réseaux
Rubrique : Domaine 3 : Graphes et réseaux
Publié le : 23 juillet 2019
Accepté le : 23 juillet 2019
Soumis le : 23 juillet 2019
Mots-clés : [INFO.INFO-SI]Computer Science [cs]/Social and Information Networks [cs.SI], [STAT.ML]Statistics [stat]/Machine Learning [stat.ML], [en] Activity prediction, Real-world networks, Link prediction, Link stream
Financement :
Source : OpenAIRE Graph- Algodiv: Recommandation algorithmique et diversité des informations du web; Financeur: French National Research Agency (ANR); Code: ANR-15-CE38-0001