Developing Predictive Simulation Framework with Confidence Level for Stochastic Process and Its Application to Knowledge Discovery

2016 Fiscal Year Final Research Report

• Project/Area Number: 26330261
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Intelligent informatics
• Research Institution: Aoyama Gakuin University
• Principal Investigator: OHARA Kouzou 青山学院大学, 理工学部, 准教授 (30294127)
• Project Period (FY): 2014-04-01 – 2017-03-31
• Keywords: 予測シミュレーション / 確率モデル / 機械学習 / 統計数学 / 知識発見

Outline of Final Research Achievements

In this work, we addressed a problem of efficiently estimating the influence of a node in information diffusion over a social network. Since the information diffusion is a stochastic process, the influence degree of a node is quantified by the expectation, which is usually obtained by very time consuming many runs of simulation. We proposed a framework for predictive simulation based on the leave-N-out cross validation technique that estimates the approximation error of the influence degree of each node without knowing the true influence degree. We experimentally showed that it can serve as a good measure to solve the problem with far fewer runs of simulation ensuring the accuracy. Besides, we applied that framework to computation of node centrality in order to show the broad utility of the proposed resampling-based framework. In addition, we also devised an efficient algorithm that runs an individual information diffusion simulation in parallel on a distributed computing environment.

Free Research Field

データマイニング、社会ネットワーク分析