Hi! I need to finish a project on sentiment analysis using SpaCy and Prophet lib
Hi! I need to finish a project on sentiment analysis using SpaCy and Prophet libraries. This is the final task:
Overview of the pipeline you developed: forecasting, ext. regressors, fine tuning NN, news sentiment analysis, evaluation with cross validation.
Results you reached as a prediction (horizon 14 days) and as a prediction of growing/decreasing of the forex
Extra models/ideas/evaluations your group have performed. (For example evaluate the model wrt a baseline, or update the series to use all data until now, or use the scraper to get more news from the past, or use the news to directly train a classifier of growing/decreasing of the next days) This is another hint for the project: Use SpaCy Projects to classify a sentiment analysis dataset composed of reddit posts, then adapt a sentiment
analysis dataset of financial news to classify news headlines instead of reddit posts. After that, use the output of
the classifier as a Prophet regressor and evaluate the impact on prediction performance. I’ll attach below the file you need to implement the coding on, and another file we used during the lab lectures with some exercises and examples.