- download corenlp from this page: https://stanfordnlp.github.io/CoreNLP/download.html
- unzip corenlp into your Communities github directory
- rename unzipped corenlp directory to ‘corenlp’
- Download stanza: pip3 install stanza -U
- Download english language: python3 -c 'import stanza; stanza.download("en")'
- pip3 install textcomplexity
- get conllu ver. of file: python3 utils/get_conllu.py data/example.json
- get analysis: txtcomplexity --input-format conllu --all-measures output/example.json.conllu > "output/example_analysis.json"
- set up your environmental variable
- on mac for example: export CORENLP_HOME='./corenlp'
- Install any python dependencies
- pip3 install nltk
- pip3 install textstat
- pip3 install vaderSentiment
- pip3 install empath
- pip3 install redditcleaner
- create separate terminal and run CoreNLP server
- cd corenlp
- java -mx4g -cp "*" edu.stanford.nlp.pipeline.StanfordCoreNLPServer -port 9001 -timeout 15000
- If you already have a process running on post 9001, you will need to stop it
- on mac: lsof -t -i tcp:9001 | xargs kill
- run code, e.g. python3 analysis/analyze_data.py
##setup Empath -Install empath
- pip install empath
-To run sample code to perform analysis use command given below. It takes file name as program argument. AnalysisUsingEmpathy.py ../data/all_text.json
##setup spacy -Install spacy
- pip install -U spacy
- python -m spacy download en_core_web_sm --To run sample code to perform analysis use command given below. It takes file name as program argument. AnalysisUsingSpacy.py ../data/all_text.json
- Stanza
- CoreNLP
- textcomplexity
- Empath
- spacy