It seems that it is possible to understand how Bitcoin behaves in the crypto market by having a look a the volume of tweets and Google Search Volume Index (SVI). As per a research conducted by the Southern Methodist University, these two values were leading price indicators for both Bitcoin (BTC) and Ethereum (ETH).
Social Media Could Predict Bitcoin Price Performance
According to this paper, researchers were able to gather data regarding Twitter mentioning Bitcoin and Ethereum, the two largest cryptocurrencies in the market. As per the report, the number of tweets and Google searches changes first before prices do.
With this analysis, it is possible to understand the role of sentiment in the market and how it plays an important role in defining virtual currency prices. Following this paper, it could be possible to understand how positive or negative people are about specific virtual currencies, the whole market and how the price of these digital assets would move.
In this study, tweet volume rather than sentiment was the key factor in predicting the price direction of digital assets. At the same time, the researchers were able to find that there are more than 21 million bots that post information about prices, advertisements and other things rather than humans discussing how virtual currencies operate.
The report explains that users were generally more positive about digital assets than on other platforms. About it, they commented:
“People who tweet about cryptocurrencies even when their prices drop have an interest in them beyond investment opportunity making the tweets biased towards positive.”
In order to analyze the data from Twitter, the researchers used the open source VADER, which stands for Valence Aware Dictionary and Sentiment Reasoner. The information that was gathered from Twitter goes back to 2014.
Google Trends data shows a high correlation between price and SVI values. The search spikes occurred before the actual increase in prices. Using machine learning the results from Google trends and Twitter were put into a linear model to verify these positive correlations.