Deep Learning: Can It Be Used To Predict Future BTC Prices?

One data scientist from India’s prestigious Vellore Institute of Technology recently outlined a method for how to predict crypto prices in real-time using a Long Short-Term Memory (LSTM) neural network. The process of predicting Bitcoin’s future price is based on deep learning – shared in a blog post published on December 2 by the researcher Abinhav Sagar.

In the post which is now viral in the Bitcoin news today, the data scientist demonstrates how to use machine learning technology to forecast prices in a sector he purported is “relatively unpredictable” as compared with the traditional markets. Sagar used deep learning to preface his demonstration – noting that while machine learning achieved some success in predicting the stock market prices, its application in the crypto field has been restricted.

In support of this claim, the scientist argued that prices of cryptocurrencies fluctuate in accordance with the fast paced technological developments and the economic, security and political factors.

The deep learning process by Sagar featured in the cryptocurrency news and the method he proposed includes the following phases:

1. Gathering data from real-time cryptocurrency movements;
2. Preparing this data for neural network training;
3. Testing the price prediction using a LSTM neural network;
4. Showing the results of his prediction.

As another software developer named Aditi Mittal outlined, LSTM is an acronym for “Long Short Term Memory” which is essentially a type of neural network that is designed to classify, process and predict time series given time lags or unknown durations.

To train his network, the data scientist (Sagar) used a dataset from real-time charts and made use of features such as price, volume and open, high and low values. He even released a link to the code for the entire project on GitHub and made use of features including price, volume and open, high and low values, etc.

Aside from market predictions, deep learning (machine learning) has been used a lot of times and has been gaining tons of traction. As reported earlier this year, NASA recently published a listing for a data scientist role, singling out the cryptocurrency and blockchain expertise as “a plus.” The agency further required qualification in one or more related fields which include machine learning, big data, IoT, analytics, statistics and cloud computing.