Indian Scientist Develops Program That Forecasts Bitcoin In Real-Time

A data scientist at India’s Vellore Institute of Technology has developed a new technique for forecasting cryptocurrency prices in real-time through the use of a Long Short-Term Memory neural system.

According to a news publication on the 2nd of December, Abinhav Sagar explained the process involved in using machine learning technology to forecast price movement in a market that is generally regarded as unpredictable.

Restriction on Price Forecast on Cryptocurrency
Sagar noted that although machine learning is a known tool in forecasting stock market prices, its utilization in the crypto space has been prohibited. To back his declaration, he added that crypto prices move at the same pace with technological developments, also with financial and political factors.

The process used in forecasting price movement include:

1- Collection of live crypto data.

2- Preparation of the collected data for neural system practicing.

3- Try out of the prediction mechanism using the LSTM neural system.

4- Assessment of the results.

Sagar made use of a dataset from CryptoCompare to train his system. He also added a link to the code for the program on GitHub and defines the objectives he used in normalizing data values in the trial for machine learning.

Sagar used a Mean Absolute Error as an assessment measure before planning and displaying the findings of the system’s forecasts. The MAE assesses the average magnitude of errors in a set of forecasts regardless of direction.

Machine Learning Starting to Pick up Steam
Aside from price forecasts, the adoption of machine learning in various sectors is starting to pick up steam.

It was announced some months ago that NASA is looking to recruit data scientists, specifically stating that those with crypto and blockchain expertise would have an edge over others. The organization went on to add that those with experience in machine learning, big data, Internet of Things, statistics and cloud computing would also be given exclusive consideration.