/nlds-PersonaBank

PersonaBank - A Corpus of Story Intention Graphs.

Version: Version 1.0. Updated May 17, 2016.

This page was migrated from https://nlds.soe.ucsc.edu/ on Sept 2023.

Overview

We present a new corpus, PersonaBank, consisting of 108 personal stories from weblogs that have been annotated with their Story Intention Graphs, a deep representation of the fabula of a story. We describe the topics of the stories and the basis of the Story Intention Graph representation, as well as the process of annotating the stories to produce the Story Intention Graphs and the challenges of adapting the tool to this new personal narrative domain. We also discuss how the corpus can be used in applications that retell the story using different styles of tellings, co-tellings, or as a content planner.

The Data

  • 108 blog stories (in original/)
  • Their corresponding .VGL (in vgl/) and Scheherazade realization files (in sch/)
  • Tutorial for blog annotation in Scheherazade (annotationTutorialForBlogs.pdf)
  • Spreadsheet describing the stories, their topics, polarity, and an excerpt of the story (personaBank.csv)
  • Spreadsheet describing the topic distribution (topics.csv)

Notes

  1. Download Scheherazade annotation software at: http://www1.cs.columbia.edu/~delson/software.shtml

  2. The original Scheherazade tutorial is available at: https://sites.google.com/site/scheherazadetutorial/

  3. Due to the nature of the Scheherazade annotation, names have been anonymized so that a name like Brian will become Jake# with trailing '#'. These '#' are only in the original story files. A .vgl loaded into Scheherazade and the Scheherazade text file will look like 'Jake' without the trailing '#'.

  4. Format of personaBank.csv. The header is:

UniqueId,Suffix,FileName,polarity,Interp layer,Topic(s),First few words

Interp layer indicates if this story has been annotation in the interpretation layer. Polarity indicates the overall polarity of the story.

  1. Format of topics.csv. The header is:
topic,parentTopic,total,posCount,posIds,negCount,negIds

If the parentTopic = -1, this is a main topic. For example:

health,-1,16,1,[108],15,[31,105,2,38,105,44,24,29,106,15,17,21,106,102]

Otherwise, it is a subtopic:

life,health,2,1,[108],1,[31]

The posIds and negIds correspond to the Uniqueid in personaBank.csv

Citation

If you use this data in your research, please refer to and cite: Lukin, Stephanie M., Bowden, Kevin, Barackman, Casey, Walker, Marilyn A. "A Corpus of Personal Narratives and Their Story Intention Graphs". In Proceedings of the 10th International Conference on Language Resources and Evaluation (LREC), Portorož, Slovenia, 2016.

Related Works

Works that use this corpus: