Mission & Concept

Unlocking the secrets of highly diverse documents stored in historical archives is crucial for advancing historical research and gaining a deeper, more nuanced understanding of the past.

The Connected History Project (CHiP), undertaken by the Masaryk Institute and Archives of the Czech Academy of Science in Prague, aims to shed light on the inner workings of the Nazi occupation in the Protectorate of Bohemia and Moravia between 1941 and 1945. CHiP seeks to unravel the complexities of this oppressive system, exploring how existing and newly established administrative structures created a well-organized network of oppression. Moreover, the project seeks to understand the profound impact of the Nazi regime on the lives of ordinary individuals.

CHiP has already amassed and is digitising a diverse collection of documents, including over 4,000 records of crimes and misdemeanors against economic regulations, more than 200 pertinent laws and regulation gazettes, and most importantly, 250 meticulously detailed case files that provide firsthand accounts of individuals who encountered conflict with the system.

The technical team of the CHiP project is employing multimodal large language models (MLLMs) to analyze these highly heterogeneous sources. These documents encompass a wide range of formats, including printed, handwritten, mixed, and in both German and Czech. The team is exploring how to effectively leverage current MLLMs and various Natural Language Processing (NLP) approaches, particularly Named Entity Recognition (NER), to analyze these documents and create automated linkages between historical sources.

Once this rich network of interconnected entities and their stories is created from our heterogeneous collection of sources, we will make them accessible through various dynamic user interfaces, aimed both towards historians and, more importantly, for educational use. The interconnected stories documented in this collection deserve to be in the virtual space.