Simple sentiment calculator for spanish texts. It calculates text positivity, negativity or neutrality by replacing words with their stored values and operating with them. It uses two types of words:
- Polarities: This words make the sentence more positive or more negative.
- Modifiers: This words modify the polarity of other words.
##Requirements
Python 2.7
pip install -r requirements.txt
Redis
##Scripts ./redis_words_loader.py -m <text_file> Loads modifiers from text_file (check modifiers.txt)
./redis_words_loader.py -p <text_file> Loads polarities from text_file (check polarities.txt)
./spell_checker.py Loads word ocurrences for the spellchecker.
./redis_load_analyser_txts.py Uses the script above to load the text files stored in /analyser/data
./run_analyser.py [-debug] Prints polarity of text
./run_analyser.py -f [-debug] Prints polarity of text from text file
./init_server.py Starts analyzer server at localhost:5000.
./make_request.py Sends a demo request to the server.
##Some links
How to Write a Spelling Corrector.
Word values in anew.txt values were calculated from Spanish ANEW. They should be removed for commercial use.
The usage of a scoring method opposed to a probabilistic one was inspired by Basic Sentiment Analysis with Python.