Quicker Literature Reviews with the Mendeley Desktop 1.11 Preview

Today we’re announcing a preview of the next release of Mendeley Desktop, which adds an experimental feature to help you explore papers and find relevant information in them more quickly. This improved reading experience is initially available for Open Access papers, but we’re hoping to expand it to additional content in future.

Table of Contents

We automatically analyze the hierarchical structure of papers, identify the main headings and present them in the ‘Contents‘ tab. From the ‘My Library’ tab, you can browse supported papers and jump directly to the section of the document you are interested in.

health-toc

Figures

We attempt to identify tables and figures that appear in the paper and list them in the ‘Summary‘ tab. You can  then select a table or figure to jump directly to the relevant section of the paper.

cancer-screen-coverage-figure

Tables of Data

Tables in the paper are located, the data is extracted from them, and the results presented in a normalised style for easier reading. You can also export the data by clicking the drop-down arrow in the header for the table and selecting ‘Copy as HTML‘. From there you can paste the data into a spreadsheet such as Excel for quick analysis or visualization.

word-segmentation-results-table

Getting the Preview

Initially, we’re making this feature available for papers in our catalog that have been identified as Open Access. To see examples of enrichments:

    1. Download the preview release of Mendeley Desktop 1.11
    2. Go to the ‘Literature Search’ pane in Mendeley Desktop, click the magnifying glass icon in the search box and select ‘Open Access’ to limit your search to open access papers.
    3. Search for a topic that interests you
    4. Browse the results. When you select a paper, if we’ve automatically extracted an outline, tables and figures then they will appear in the ‘Contents’ and ‘Summary’ tabs in the right-hand pane.

Feedback

This is a new and experimental feature which we’re making available for early feedback. We know that we have plenty of work to do to expand coverage to more papers and improve the recall and accuracy of extraction, especially for tables with more complex layouts.

The research that enabled this feature was developed as part of the EU-funded CODE project, with partners at the University of Passau and the Know-Center in Graz.

Join Us!

Would you like to help us make the lives of researchers easier? Interested in developing algorithms for data extraction, working with a library of the world’s research or creating beautiful apps for scientists? We’re looking for a data scientist to help us extract information from papers and engineers to help bring the fruits of their work to users on desktops , mobile devices and the web. If you’d like to be involved – please get in touch! For examples of the R&D work we’ve been doing internally and with univeristies, see the Mendeley profiles of Roman and Kris.