Big Data final project

factors (including text analysis)

factors from tushare & uqer API emotion and heat index from uqer API CCTV (and other text data) from tushare API visualization

machine learning models for stock prediction

the model integrate machine learning models (logistic regression, random forest, gradient boosting, and SVM) with multi factors and an overall index generating from CCTV news.

• classification model to find stocks beating HS300 index
• regression model to find stocks with highest return