Development of Interactive Advisory System for Security Export Control

2022 Fiscal Year Final Research Report

• Project/Area Number: 20K12556
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 90020:Library and information science, humanistic and social informatics-related
• Research Institution: Hokkaido University
• Principal Investigator: OBAYASHI AKIHIKO 北海道大学, 産学・地域協働推進機構, 教授 (80798124)
• Co-Investigator(Kenkyū-buntansha): RZEPKA Rafal 北海道大学, 情報科学研究院, 助教 (80396316)
• Project Period (FY): 2020-04-01 – 2023-03-31
• Keywords: export control / trade security / expert system / dialog system

Outline of Final Research Achievements

We have designed a chatbot system to assist academics dealing with research output or technology subject to sensitive export control regulations. Due to a lack of experts in this field, it was crucial to develop AI technology that can ease their burden. Our solution was an interactive chat system that guides users to relevant sections of export control regulatory texts. It incorporates several features: publication uploading, synonym queries, CAS number retrieval, references to EU documents, and automated translation for foreign researchers. By expanding the system's glossary and leveraging natural language processing, it can handle non-technical queries unrelated to legal terms. While our primary objective was to create a functional system, we also conducted experiments, which are detailed in scientific publications.

Free Research Field

Security Export Control in Academia

Academic Significance and Societal Importance of the Research Achievements

Our system aims to reduce expert workload and enhance researchers' awareness of often overlooked dangers. This is vital as the academic world is susceptible to potential threats to national security. Additionally, we demonstrated the unreliability of relying solely on neural methods.