Using Sequence Constraints for Modelling Network Interactions

• SMEDT Johannes De

  The University of Edinburgh

• MORI Junichiro

  The University of Tokyo

• OCHI Masanao

  The University of Tokyo

抄録

The ubiquitous nature of networks has led a vast number of works dedicated to the study of capturing their information. Various graph-based techniques exist that report on the characteristics of nodes and edges, e.g., author-citation networks, social interactions, and so on. A significant amount of information can be extracted by summarizing the surrounding network structure of nodes, e.g., by capturing motives, or walk patterns. In this work, we present a new way of capturing the interaction between nodes in a network by making use of the sequence in which they occur. (1) The objective of this paper is to make use of behavioural constraint patterns; a concise but detailed report of node's interactions can be constructed that can be used for various purposes. (2) It is shown how the constraint patterns can be mined form interaction data, and how they can be used for various applications.

収録刊行物

• 人工知能学会全国大会論文集

  人工知能学会全国大会論文集 JSAI2019 (0), 1K4E104-1K4E104, 2019

  一般社団法人 人工知能学会

キーワード

• Sequence Constraint
• Behavioural Constraint
• Network Analysis

詳細情報

• CRID: 1390001277395333248
• NII論文ID: 130007658367
• DOI: 10.11517/pjsai.jsai2019.0_1k4e104
• 本文言語コード: en
• データソース種別:
  • JaLC
  • CiNii Articles
• 抄録ライセンスフラグ: 使用不可