A digital library is a 24×7 virtual knowledge repository of millions of books and publications. The contents are sourced from educational boards, institutions, publishers, online video tutorials etc. and hence are formatted differently.
Library Scientists needed to capture the data and metadata in a uniform format to enable proper searching among the millions of available resources. Given the volume of data, the semi-manual effort was not scalable and error prone.
Alumnus automated the entire data acquisition process. This included rule engines to facilitate mapping, curation, content stitching and post-processing. A portal for role-based access was also built. This provided library scientists, who are not programmers, with drag-and-drop controls and other intuitive ways of defining mapping and curation rules.
This significantly improved traceability and manageability in the process. It resulted in improved efficiency and reliability of data capture.
Technology Stack:
Java
