Peer Review

[XinranPan]

Things Liked:
1.This is a cool idea! It definitely provides a different perspective on market predictions and to my knowledge, it is somewhat unique as traditional forecasting on the market only includes hard numbers.
2. Using NLP to classify and identify the words is a practical idea and is related to the material we'll be covering this semester.
3. I like that you incorporate the NLP method of public sentiment and take it into account for creating a new portfolio with the additional given information.

Areas for Improvement:

1. Can do a better job of describing the data set. Does the data set from twitter contain tweets related to stocks? What kind of data exists, any missing inputs, etc.
2. What is the difference between public and market sentiment? Would be helpful to provide your definition of both sentiments, especially public sentiments. Are they general feelings and emotions people have when it comes to the stock market?
3. Would like to see more details on the different sentiment categories and thought as to which words and phrases determine a particular mood.