Technical and sentiment analysis to predict the stock market with machine learning models based on historical time series data and news article sentiment collected using APIs and web scraping.
The notebook named "Deep Learning: Long Short-Term Memory (LSTM)" has indeed a wrong logic and the results are not realistic. The problem is you created the test data and fed the model to predict the future. Well, that's the problem! You don't know the future and you can't use the test set to make the predictions. Instead, you have the first 60 data points that you use to predict the 61th. Then you use the second 60 data points (including the one you just predicted) to predict the 62th and so on. The current implementation is wrong.
Hello Alison, I have a problem. Do you notice that the error function is too big? The evaluation of the model is bad, but when I draw the test and prediction data, the shape is good. Are there suggestions to improve the results or is this the maximum possible?
Hi, I am facing problems in installing packages regarding web scraping and I need to work on the files, they are very wonderful. Thank you for that great work. Can you attach all the csv and Pkl files thatit usre in the project so that I can work on them initially