ChatGPT-GNN Dataset

The data can be found in the Data folder, which contains two files:

• ticker_train_data.json: This file holds the data utilized for training and validation of our model.
• ticker_test_data.json: This file contains the data used for model evaluation.

To load the data, you can start with 4 lines of code:

import pandas as pd
import json
train_data = pd.read_json('./Data/ticker_train_data.json')
test_data = pd.read_json('./Data/ticker_test_data.json')

The Affected Companies column provides two key insights:

• Companies that ChatGPT predicts will be influenced by the financial news.
• The sentiment indicating the nature of the impact on these companies (e.g., positive or negative).