[ad_1]
According to a research article published by Southern Methodist University, the volume of tweets and the Google Search Volume Index (SVI) have proven to be the key price indicators for Bitcoin and Ethereum.
The importance of feeling
In the document, researchers collected data on Twitter mentioning Bitcoin and Ethereum; the same thing was done using Google's trends. Based on previous research ideas, the hypothesis was that the number of tweets and their feelings (positive and negative) could influence prices. In the study, it was discovered that the number of tweets and Google searches was changing before prices.
The role of sentiment in technical or market analysis is to reveal people's attitudes towards a market or index (in this case, Bitcoin and Ethereum). The sentiment analysis theory is a branch of technical analysis that states that prices are discounted, and that price trends ultimately reflect the psychology of the crowd.
So, theoretically, if you could measure the positive or negative impact of people's shared views on a particular stock or cryptocurrency, you could evaluate the price trajectory.
Although, in this particular study, the volume of tweets, not feelings, is a determining factor in the price of cryptocurrencies. The absence of sentiment being the main factor has been theorized due to the amount of "noise" on Twitter regarding currencies compared to the actual conversation.
For example, researchers found that 21 million bots on Twitter mostly published factual information about prices, advertisements, spam, and so on. Users do not really discuss what they think of Bitcoin or Ethereum.
The researchers also found with Twitter that sentiment was generally positive, even when the prices of Bitcoin and Ethereum were down.
People who tweet about crypto-currencies even when their prices go down have an interest in them, beyond the possibilities of investment, which makes the tweets biased.
Despite their findings, researchers did not completely rule out sentiment analysis using different modeling techniques.
Methodology
In this study, researchers opened VADER (Valencia Knowledge Dictionary and Sense Reasoner) for tweet data analysis. The tweet data goes back to 2014 on the website bitinfocharts.com. Google Trend Data (SVI) dates back to 2004, in proportion to all searches on all Bitcoin and Ethereum terms.
Results
For Google's trend data, the report found that the price was strongly correlated to searches for the Bitcoin and Ethereum keyword, and that these research peaks had occurred before the actual price increase was observed.
Another strong correlation between Twitter and the price of Bitcoin was found, except that this time with more convincing results.
Finally, using machine learning, the results of Google trends and tweet data were also integrated into a linear model to verify positive correlations. The data was divided between a training model and the tests between 80% and 20%.
Social media helps monitor investors' chatter
VADER data could provide valuable data for investors to measure market sentiment.
Previously, Bitcoinist has already highlighted the importance of social media chat on Twitter with tools such as the Twitter Hype Index. This is the first time Twitter and IVR data show that the most popular cryptocurrency prices are not tracked.
Can social media data provide valuable business information? Share your thoughts below!
Pictures via Shutterstock
[ad_2]
Source link