Incomplete upload

The machine has been trained to parse PDFs of scientific articles/patents to index their metadata, incl. abstracts. However, some PDFs can have an inconsistent format or labeling, for example being in pure image format or in a poor OCR version, which can cause difficulties in processing the document.

The PDF article should be also published in English, or at least have an English title and abstract.

When uploading a Document Collection (.csv, .bib, .nbib, .xml) all documents need to have a title and abstract in order to be processed and added to your content collection.

Indexing of formats other than PDFs of scientific articles/patents, e.g. ppts, Word documents or .txt files, is currently not supported.