Effectively preserving access to digital content over time is dependent on availability of an appropriate IT infrastructure including access to appropriate rendering software and its requisite operating systems and hardware. The complexity of this task increases over time and with the size and heterogeneity of digital collections. Automating notifications on file format endangerment and decision recommendations can greatly improve preservation planning processes. This paper presents work in progress that contributes to the design and testing of an automated file format endangerment notification and recommendation system. This system’s design is based on concepts explored in previous research, but it presents the novel application of statistically generated similarity profiles and machine-generated recommendations based on human expert input.

File format endangerment, institutional risk profiles, recommender systems, notification systems

application/pdf

Proceedings of the 12th International Conference on Digital Preservation

School of Information and Library Science, University of North Carolina at Chapel Hill