Expanding KTH's Canvas ecosystem to support additional automated services : Automating the injection of theses and their metadata into a digital archive

Abstract: Whenever a student submits their final version of their thesis, a series of processes is triggered to finalize and archive the report. These processes are often handled in a less than efficient way which results in excessive manual labor and costs that can be prevent if automated. This report describes a solution that automates the series of processes that occur following a final thesis report submission. By utilizing the available information in a Canvas course and the content in the submitted thesis much of the manual cut-and-paste effort is avoided. Entering this data into DiVA is done by automated interaction via a browser, as DiVA does not have an application programming interface that could be used. The conclusion is that it is possible to automate this process through a headless browser. However, the automated parsing of the PDF version of the thesis proven to be inconsistent which results in the extracted data being inconsistent. With some improvements to the parsing module, the entire process could be fully automated.