This issue continues the work from #54. Use the novelty scores derived from OpenAlex data to:
Check papers ranked as particularly novel or common to gauge if the results make sense
Decide which topic level we should focus on using
Write a pipeline for generate the topic-level novelty scores
Explore charts that could be used for the dashboard
Topic novelty: Document's novelty measure, aggregated at topic level, to gauge which topics are associated with more novel research (let's say, five most recent years)
Topic novelty dynamics across years: Time series and/or growth vs magnitude plots
This issue continues the work from #54. Use the novelty scores derived from OpenAlex data to: