Understanding the spatio-temporal dynamics of archaeological sites across the Japanese archipelago using natural language processing

2023 Fiscal Year Final Research Report

• Project/Area Number: 21K21323
• Research Category: Grant-in-Aid for Research Activity Start-up
• Allocation Type: Multi-year Fund
• Review Section: 1002:Human informatics, applied informatics and related fields
• Research Institution: Osaka Institute of Technology
• Principal Investigator: Fumihiro Sakahira 大阪工業大学, 情報科学部, 准教授 (70578129)
• Project Period (FY): 2021-08-30 – 2024-03-31
• Keywords: 自然言語処理 / 考古遺跡 / 発掘調査報告書

Outline of Final Research Achievements

This study developed a method to extract culturally related archaeological sites with high similarity by applying natural language processing (NLP) techniques to excavation reports and calculating the similarity between these reports. Specifically, NLP was applied to the excavation reports of sites classified as similar in existing archaeological research papers, and their similarity was calculated to verify the consistency with the classifications in the existing research papers. As a result, it was found that this approach is effective for excavation reports, which often contain many variations in the notation of technical terms and have not yet been sufficiently developed as a corpus.

Free Research Field

計算考古学

Academic Significance and Societal Importance of the Research Achievements

本手法を開発することで考古学研究者は、遺物や遺構など複雑な要素で構成される文化の伝播を研究する際に、従来のように発掘調査報告書を1冊ずつ虱潰しに読んだうえで必要な情報かどうかを判断するという多大な労力から解放されることで、容易に長期間の広域における大局的な展開の把握が可能となる。さらに、従来では見落としていた遺跡の情報が抽出される場合もあり、再発見や再解釈につながることが期待できる。さらに、この成果は個別の研究のみならず、考古学の研究方法の変革や学術文書検索システムのアルゴリズム改良にも貢献することが期待できる。