Dogecoin price movement is often considered to be influenced by Twitter sentiment, particularly from so-called "Whales", or influential people, such as Elon Musk.
Through data collection and machine learning, we can aggregate twitter posts within a specific time frame and analyze the sentiment of those posts algorithmically.
I randomly scraped 5000 tweets posted between June 26th and June 28th that had the word or tag "dogecoin" in them. I then imported the tweets into a DataFrame in Python for analysis.
There are a variety of word and sentence-based tokenizers that machine learning practitioners use to prepare the text for machine learning models, but for time's sake, I used MonkeyLearn's API, which allows users to quickly analyze text sentiment using MonkeyLearn's pre-made text models.
The 5000 tweets in aggregate were classified as positive with a 93.3% confidence level.
This leads me to believe that over the past few days, the general Twitter sentiment around Dogecoin has been relatively positive. Of course, this is a very limited analysis, and more data and more specifically tuned models would be required to get a very accurate picture of Twitter sentiment. Furthermore, the extent to which Twitter sentiment influences future Dogecoin price movement is uncertain but certainly limited, especially when looking at Twitter posts in general rather than focusing on "Whales".