All Sci Fi novels published since 1900 were scraped from Good Reads, and for each novel, all reader comments, plot descriptions, and user-generated tags were compiled. Keywords and concepts were added to each novel by parsing the text described above and mapping them to a curated dictionary of SciFi keywords and concepts. 2,633 novels published since 1990 had at least 50 reviews and contained at least one keyword from the keyword corpus.
After keyword enhancement, a book network was created by linking novels if they share similar keyword. Network clusters identify Keyword Themes - or groups of similar books that are labeled most commonly shared keywords in the group.
The network was generated using the open source python 'tag2network' package, and visualized here using 'openmappr''. The scripts for analyzing this dataset are available at https://github.com/ericberlow/SciFi
This is a collaborative project of Bethanie Maples, Srini Kadamati, and Eric Berlow
NOTE - This visualization performs best in Chrome and Safari browsers full screen - and is not optimized for the small screens of mobile devices.
How to Navigate this Network:
Click on any node to to see more details about it.
Click the whitespace or 'Reset' button to clear any selection.
Use the Snapshots panel to navigate between views.
Use the Filters panel to select nodes by any combination of attributes.
Click the 'Subset' button to restrict the data to the selected nodes - The Filters panel will then show a summary of that subset.
Use the List panel to see a sortable list of any nodes selected or subset. You can also browse their details one by one by clicking on them in the list.