What features could be useful for predicting prices?

[mltrev23]

We are currently using financial factors to predict S&P 500 price now.
But we might potentially need to include sentiment data for prediction.
Does anyone have an idea for this?
What models can we use and how to combine these with original features?

[drunest]

Add Sentiment as New Columns:

Calculate a daily sentiment score (e.g., average sentiment).
Add it as a new column to your financial dataset.
Train Your Model:

Use a straightforward model like Linear Regression or XGBoost with the combined data (financial + sentiment features).
Evaluate Results:

Check if adding the sentiment data improves your prediction accuracy.
Basically, just treat sentiment data as extra input columns alongside your financial factors, and train your model as usual.