GitHub - charlesbueso/svmStockSentiment: Simple support vector classifier implemented with scikit-learn for classifying positive or negative sentiment associated with the Dow Jones index, using binary label to predict whether the market rose or fell on any given specific date.

charlesbueso / svmStockSentiment Public

Notifications You must be signed in to change notification settings
Fork 0
Star 1

Simple support vector classifier implemented with scikit-learn for classifying positive or negative sentiment associated with the Dow Jones index, using binary label to predict whether the market rose or fell on any given specific date.

1 star 0 forks Branches Tags Activity

Star

Notifications

Branches Tags

Name		Name	Last commit message	Last commit date
Latest commit History 7 Commits
.ipynb_checkpoints		.ipynb_checkpoints
Combined_News_DJIA.csv		Combined_News_DJIA.csv
README.txt		README.txt
stockSentimentSVM.py		stockSentimentSVM.py
stopWords.txt		stopWords.txt

Repository files navigation

----------------------- R E A D M E ----------------------------

------------- BEFORE RUNNING:
To run, the following text files need to be on source folder of the project:
       - Combined_News_DJIA.csv
       - stopWords.txt

------------- REFERENCES:
Sun, J. (2016, August). Daily News for Stock Market Prediction, Version 1. 
Retrieved [March 9th, 2022] from https://www.kaggle.com/aaron7sun/stocknews.


Stop words extracted from:
    https://www.ranks.nl/stopwords

-------------- NOTES:

Data (news headlines and DJIA 1 or 0 feature) was split as follows:
1989 (70% training, 15% development, 15% testing)
1392 training
298 development
299 testing

----------------------- R E A D M E ----------------------------