Prospective Analysis of Large and Complex Partially Observed Temporal Social Networks

Abstract

The objective of our study was to explore in detail the hypothesis that a unified approach is feasible for modeling large-scale temporal social networks that: provide predictions and uncertainty estimates for the network structure and the properties of nodes, capture simultaneously positive and negative influences among nodes, consider simultaneously node properties of multiple types and of different temporal dynamics, address multiple kinds of interactions, allow learning in partially observed temporal graphs including predicting properties for emerging nodes. The objective of the study is completely achieved, and the results are published in 52 articles listed at the end of the report.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Aug 14, 2018
Accession Number
AD1085877

Entities

People

  • Zoran Obradović

Tags

Communities of Interest

  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Aneurysm
  • Artificial Intelligence
  • Bayesian Networks
  • Computational Science
  • Computers
  • Crime
  • Data Mining
  • Health Services
  • Information Processing
  • Information Science
  • Information Systems
  • Knowledge Management
  • Machine Learning
  • Monte Carlo Method
  • Network Science
  • Neural Networks
  • Probabilistic Models

Readers

  • Computational Modeling and Simulation
  • Neural Network Machine Learning.