Influencer-Generated Reference Groups

Code and data to accompany Influencer-Generated Reference Groups by Junque de Fortuny and Lee 2021, t.b.d.

[TOC]

Tools

Our code relies on the following tools which must be installed prior to running the code.

Natural Language Processing

Name	URL
NLTK wordnet lemmatizer	https://www.nltk.org/howto/wordnet.html
LIWC	http://liwc.wpengine.com/
SentiWordNet	https://github.com/aesuli/SentiWordNet
General Inquirer (GI)	http://www.wjh.harvard.edu/~inquirer/
ANEW	https://csea.phhp.ufl.edu/media/anewmessage.html
Sentic	https://sentic.net/
Evaluative Lexicon	http://www.evaluativelexicon.com/

Machine Learning

Clustering

Name	URL
Weighted Dirichlet Prior GMM	https://scikit-learn.org/stable/modules/mixture.html https://github.com/ktrapeznikov/dpgmm

Embedding

Name	URL
Counter-fitting	https://github.com/nmrksic/counter-fittingNLTK
Numberbatch Embeddings	https://github.com/commonsense/conceptnet-numberbatch
ConceptNet	https://conceptnet.io/
UMAP	https://umap-learn.readthedocs.io/en/latest/

Image processing

Name	URL
DISTS	https://github.com/dingkeyan93/DISTS
BRISQUE (alternative image quality analysis)	https://github.com/ocampor/notebooks/blob/master/notebooks/image/quality/brisque.ipynb

Code/Set-up

1. Clone the github repository
2. Tools need to be installed in the src/lib folder
3. Data should be placed in the src/data folder
4. An empty src/output folder should be created.
5. Run by opening Main.ipynb