Skip to contents

A stratified subsample of corpus_scifi_pulp: 4,800 passages of approximately 350 words each, drawn from 2,072 valid pulp magazine issues (1926-2017) after dropping empty and OCR-corrupted issues. Issues were split into non-overlapping passages, then sampled across magazine-by-decade cells (floor of 30, cap of 150 per cell). The doc_id column links each passage back to the source issue in corpus_scifi_pulp.

Format

A data frame with 4800 rows and 7 variables.

Source

https://archive.org/details/pulpmagazinearchive

Variables

  • passage_id. Unique identifier for each passage

  • doc_id. Identifier of the source issue (links to corpus_scifi_pulp)

  • magazine. Title of the magazine

  • year. Year the source issue was published

  • decade. Decade of publication

  • passage_index. Position of the passage within the source issue

  • passage_text. Passage text (de-hyphenated, OCR line-wrap breaks joined)