Hype and Diversity - PubMed dataset

This dataset captures ‘Hype’ and 'Diversity', including article-level (pmid) and author-level (auid) data within biomedical abstracts sourced from PubMed. The selection chosen is ‘journal articles’ written in English, published between 1991 and 2014, totaling 421,580 (merged_df).
The classification of hype relies on the presence of specific candidate ‘hype words’ and their abstract location. Therefore, each article (PMID) might have multiple instances in the dataset due to the presence of multiple hype words in different abstract sentences. Diversity is classified for ethnicity, gender, academic age, and topical expertise for authors based on the Rao-Sterling Diversity index.
File1: merged_auids.csv (Important columns defined)
• AUID: a unique ID for each author
• Genni: gender prediction
• Ethnea: ethnicity prediction
#################################################
File2: merged_df.csv (Important columns defined)
- pmid: unique paper
- auid: all unique auids (author-name unique identification)
- year: Year of paper publication
- no_authors: Author count
- journal: Journal name
- years: first year of publication for every author
- Country-temporal: Country of affiliation for every author
- h_index: Journal h-index
- TimeNovelty: Paper Time novelty
- nih_funded: Binary variable indicating funding for any author
- prior_cites_mean: Mean of all authors’ prior citation rate
- insti_impact: All unique institutions’ citation rate
- mesh_vals: Top MeSH values for every author of that paper
- hype_word: Candidate hype word, such as ‘novel'
- hype_value: Propensity of hype based on the hype word, the sentence, and the abstract location
- hype_percentile: Abstract relative position of hype word
- relative_citation_ratio: RCR