Описание

Stock price prediction is a tedious task because the changes in stock prices do not follow any specific pattern and purely depend on supply and demand over a period of time. Machine learning can identify these patterns by using algorithms to learn specific characteristics of stock prices. One of the most well-known networks for time series prediction is the long short-term memory (LSTM), a recurrent neural network (RNN) that can remember information over a long period of time, making it very useful for stock price prediction. RNNs are well suited for time series data and can maintain an internal state that caches a summarized version of the information seen so far by processing the data step by step.

Презентация проекта