Digitization of historical records has produced a significant amount of data for analysis and interpretation. A critical challenge is the ability to relate historical information across different archives to allow for the data to be framed in the appropriate historical context. This paper presents a real-world case study on historical information integration and record matching with the goal to improve the historical value of archives containing data in the period 1800 to 1920. The archives contain unique information about M\'etis and Indigenous people in Canada and interactions with European settlers. The archives contain thousands of records that have increased relevance when relationships and interconnections are discovered. The contribution is a record linking approach suitable for historical archives and an evaluation of its effectiveness. Experimental results demonstrate potential for discovering historical linkage with high precision enabling new historical discoveries.
翻译:历史记录的数字化产生了大量可供分析和解读的数据。一个关键挑战在于如何关联不同档案中的历史信息,以便在适当的历史背景下构建数据框架。本文提出了一项关于历史信息整合与记录匹配的真实案例研究,旨在提升1800年至1920年间档案数据的历史价值。这些档案包含了关于加拿大梅蒂斯人及土著人群体的独特信息,以及他们与欧洲定居者的互动记录。档案中数千条记录在发现关联与相互联系后具有更高的研究意义。本文贡献在于提出一种适用于历史档案的记录链接方法,并评估其有效性。实验结果表明,该方法能够以高精度发现历史关联,从而推动新的历史发现。