screenplay_to_events: Tag a well-formatted screenplay in PDF format and extract...

In pj398/charinet: A Package for Character Interaction Network Analysis

Tag a well-formatted screenplay in PDF format and extract event list

Description

This function reads in a well-formatted screenplay (via pdftools::pdf_text()), parses it and tags each line based on a series of regular expressions. The function attempts to tag each line as either a scene boundary, scene description, character name, dialogue, dialogue description, stage direction, or page information. These tags are then used to generate an event list of character interactions based on these tags. In particular, the scene boundary tags and character name tags are used to identify which characters speak within scenes.

In order to work, the screenplay must use consistent patterns of indentation for different components and follow certain industry standard formatting conventions for screenplays. It is unlikely to work if the screenplay is watermarked. See Agarwal et al. (2014) for more discussion of the limitations of regex-based tagging of screenplays.

Usage

screenplay_to_events(pdf_file, window = 5)

Arguments

pdf_file	File path to the screenplay PDF file.
window	A numeric value specifying the context window within which to look for 'recipients' of a line of dialogue. For a given line of dialogue t spoken by i, the function will consider another character j to be a 'recipient' of i's dialogue if j spoke within the same scene and within n=window lines of dialogue of t.

Value

A matrix containing a time-ordered multicast event list. The first column contains an event index, the second contains a scene index, the third contains the speaker ID, and the remaining columns contain dummy variables for each character indicating whether they were the 'recipient' of the line of dialogue (which is determined by whether they spoke within the same scene and within n=window lines of dialogue of the current line).

References

Agarwal, Apoorv, Sriramkumar Balasubramanian, Jiehan Zheng, and Sarthak Dash. ”Parsing Screenplays for Extracting Social Networks from Movies.” In Proceedings of the 3rd Workshop on Computational Linguistics for Literature (CLfL), 50-58. Gothenburg, Sweden: Association for Computational Linguistics, 2014.

Examples

## Not run:  my_pdf <- "path/to/pdf/of/screenplay.pdf"
my_event_list <- screenplay_to_events(my_pdf) 
## End(Not run)

pj398/charinet documentation built on May 2, 2024, 10:28 p.m.