Design Theory of Probabilistic Computational Models for Community Detections based on Non-Additive Volume and Entropy

2016 Fiscal Year Final Research Report

• Project/Area Number: 25280089
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Partial Multi-year Fund
• Section: 一般
• Research Field: Soft computing
• Research Institution: Tohoku University
• Principal Investigator: TANAKA Kazuyuki 東北大学, 情報科学研究科, 教授 (80217017)
• Co-Investigator(Kenkyū-buntansha): 安田 宗樹 山形大学, 理工学研究科, 准教授 (20532774) ; 和泉 勇治 東北大学, 情報科学研究科, 准教授 (90333872) ; 片岡 駿 東北大学, 情報科学研究科, 助教 (50737278)
• Co-Investigator(Renkei-kenkyūsha): KINOSHITA KENGO 東北大学, 大学院情報科学研究科, 教授 (60332293)
• Research Collaborator: Hsu Chiou-Ting 台湾国立清華大学, コンピュータ科学部, 教授 ; Furtlehner Cyril フランス国立情報学自動制御研究所, 研究員 ; Zdeboraba Lenka サクレー原子力庁センター, 研究員 ; Ricci-Tersenghi Federico ローマ大学ラ・サピエンツァ校, 物理学科, 准教授 ; Welling Max アムステルダム大学, 理学部, 教授 ; Zhang Pan 中国科学院理論物理研究所, 准教授
• Project Period (FY): 2013-04-01 – 2017-03-31
• Keywords: 数理工学 / 機械学習 / 統計数学 / 情報統計力学 / 確率的情報処理

Outline of Final Research Achievements

We have proposed a new algorithm to detect the community structure in a network that utilizes both the network structure and vertex attribute data. The problem addressed this paper is the detection of the community structure from the information of both the network structure and the vertex attribute data. The detection of the community structure in our method is achieved by using belief propagation and an EM algorithm. We numerically verified the performance of our method using computer-generated networks and real world networks. Actually, a part of our research results and aspects in complex networks are expanded to the traffic data reconstruction problem and on-demand color calibration system to track pedestrians across non overlapping fields of fixed camera view. Moreover, we have proposed a new quantum annealing method to achieve the optimization of modularity in community detection problems.

Free Research Field

確率的情報処理