Lars Wieneke

HistoGraph -- A Visualization Tool for Collaborative Analysis of Networks from Historical Social Multimedia Collections

This paper describes the design and development of histoGraph, an interactive tool for explorative visualization and collaborative investigation of historical social networks from multimedia collections. Developed in an interdisciplinary collaboration of computer scientists, historians, HCI researchers and interface designers, the tool aims at supporting historians in the discovery and historical analysis of relationships between people, places and events. A special focus is on the identification and interactive visualization of social relations from historical photo collections through a combination of automatic analysis and expert-based crowd sourcing. The tool design bridges the gap between established network analysis and visualization techniques and traditional hermeneutic research methods in historical research. It integrates visual exploration with hybrid social graph construction, hypothesis formulation and the consultation of digitized primary sources. A formative evaluation of the current prototype, developed as a domain-specific application for historians in the field of European integration points to opportunities and critical factors in applying this approach to support and further current research practices in digital humanities.

histoGraph as a Demonstrator for Domain Specific Challenges to Crowd-Sourcing

histoGraph provides an integrated pipeline for the extraction of co-occurrence information in historical photos to build an exploreable social graph of relationships that can lead to new insights for historical research. The application leverages on the CUbRIK platform for human/machine computation and applies a hybrid approach to face-detection and -recognition that combines the strengths of algorithmic analysis with expert and generic crowd sourcing. Following a general overview of our approach, we explore the surplus value of human touch for the identification of identities in historical image collections through a uniform crowd-sourcing approach. We find that only a combination of generic and expert crowds yields promising results. Even though the application was designed and developed for a specific target audience, we aim not only at demonstrating the current functionality but also identify and discuss several core principles that can be transferred to other domains.