This repo contains the scripts to clean text from CHI papers (from PDF extraction); and to extract modifiers about interaction. The repo does not contain the actual data itself, why you need to implement get_proceedings
in proceedings.py
, such that it returns the raw texts from the papers you want to analyze.
Developed for Python 2.7; install required packages using pip install -r requirements.txt
Tools for turning extracted text from CHI PDFs into cleaned strings to perform analysis. The tools remove ill-formed OCR, lexical optimizations, and remove headers, author info and other meta data from strings.
See cleantext/cleantext.py
for how to clean your own extracted CHI PDFs
Run tests:
python -m cleantext.test
Extracting 'interaction' modifiers. Implement get_proceedings
in proceedings.py
first.
Run it:
python -m modifiers.noun_phrases
Citation:
@article{Hornbaek:2019:WMX:3341168.3325285,
author = {Hornb{\ae}k, Kasper and Mottelson, Aske and Knibbe, Jarrod and Vogel, Daniel},
title = {What Do We Mean by ``Interaction''? An Analysis of 35 Years of CHI},
journal = {ACM Trans. Comput.-Hum. Interact.},
issue_date = {July 2019},
volume = {26},
number = {4},
month = jul,
year = {2019},
issn = {1073-0516},
pages = {27:1--27:30},
articleno = {27},
numpages = {30},
url = {http://doi.acm.org/10.1145/3325285},
doi = {10.1145/3325285},
acmid = {3325285},
publisher = {ACM},
address = {New York, NY, USA},
keywords = {Interaction. history, quality, style},
}