Surely 9.8 million tweets can’t be wrong. Computer scientists analyzed millions of tweets from 2008, characterizing trends in “mood dimensions” with evocative  names such as “Calm,” “Alert,” “Vital,” “Sure,” “Kind,” and “Happy.” The researchers found that the “Calm” index tracked well with the ups and downs of the Dow Jones Industrial Average, while preceding them by three or four days.

In other words, the twitter analysis appears to be a predictor of which way the stock market is going, along with the magnitude of the swings.

Graph of twitter analysis overlaid on stock market Dow Jones.This chart shows the “Calm” twitter analysis when three days, overlaid on the Dow. The artificial intelligence sentiment analysis techniques used to analyze the tweet data are predicting changes in the Dow with an accuracy of 87.6%.

Here’s the story, entitled “Twitter Predicts the Stock Market” with analysis by Aaron Saenz for singularity.com.

Tags: , , ,

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Share