Modulus white paper: quantification of social media sentiment aids cryptocurrency forecasting abilities

 

Amid global fallout from the Covid-19 pandemic and the uncertainty surrounding the 2020 presidential election, cryptocurrencies have spiked, surpassing even their 2017 highs. Richard Gardner, CEO of Modulus, a US-based developer of ultra-high-performance trading and surveillance technology that powers global equities, derivatives, and digital asset exchanges, including a white-label security token exchange solution, offered insight into how social media can be used to enhance trading behavior.

“We do more than just fintech, and we’re much more than just a cryptocurrency exchange provider,” explained Gardner. “Modulus actually developed world’s largest, real-time social media sentiment analysis system. Later, we successfully adapted it to analyze stocks and real estate.”

Essentially, in each social media input (a tweet, for example), the sentiment analysis engine obtains the data before artificial intelligence is used to quantify that data, assigning it an emotion-based representation, such as anger and sadness. Alternatively, researchers can assign each tweet a discrete value on a spectrum, such as a numeric value system stretching from 1-5, which would indicate negative, neutral or positive sentiments.

“There’s no question that information from social media can be used to anticipate cryptocurrency value. Consider the case of John McAfee, who was once a celebrity endorser, charging large sums of money to endorse tokens in the cryptocurrency space. Token start-ups knew that his followers would create a buzz around tokens he mentioned

and that the buzz would boost, at least in the short-term, the token’s price,” noted Gardner.

As a security token exchange solution and cryptocurrency exchange provider, Modulus developed a white paper that traders can use as a roadmap to begin incorporating social media into their research routine. “We’ve developed formulas, based on peer-reviewed scientific research — it accounts for the set of users who follow the original message creator, those who liked it, and those who shared it. That allows us to quantify and score the audience which each sentiment receives,” explained Gardner. “To quantify the trend of sentiment surrounding cryptocurrency, we also include the linear momentum of sentiment. We are particularly interested in the rate of change in sentiment.”

Both the final sentiment score, as well as the quantified momentum are fed into the Modulus Sentiment Analysis Engine, and technical indicators are developed. “Through our experience using artificial neural networks for price forecasting, without social media data, we found that using relatively small training datasets led to more accurate short-term predictions. However, we also believe that the addition of sentiment information, especially our measure of momentum, will help our models identify indicators for sudden, large price fluctuations that the market has struggled to capture until this point,” said Gardner.

Modulus is known throughout the financial technology segment as a leader in the development of ultra-high frequency trading systems and exchanges. Over the past twenty years, the company has built a client list which includes NASDAQ, Goldman Sachs, Merrill Lynch, JP Morgan Chase, Bank of America, Barclays, NASA, Siemens, Shell, Yahoo!, Microsoft, Cornell University, and the University of Chicago. Since 2018, Modulus has offered a cutting-edge security token exchange solution and is widely considered to be the most technologically advanced white label cryptocurrency exchange provider in business today

Leave a reply

Please enter your comment!
Please enter your name here