(Update: I just added our analysis of the early sentiment toward the Star Trek movie here.)
In anticipation of the “Mining Twitter” session I am co-presenting with Warren Sukernek of Radian6 at the Twitter conference, Warren sent me a batch of the 1500 earliest Tweets mentioning the Kindle DX to run through our sentiment analysis engine. I filtered them down to a set that encompassed 90% of total reach (based on followership) and ran them through Sentimine (basic pass.) The resulting data is charted below. Looks like the jury is still out, (the tweets are scored overwhelmingly neutral) but the response definitely slants positive so far. Ratio of positive to negative based on qty/reach is 1.86/1.51.
We’ll expand on this this as the product lands in people’s hands. Note that we’ll have a more comprehensive report to show at the conference that analyses the Kindle 2. We’ll not only show how well the product is liked, we’ll have an analysis of why it was liked/not.
This reflects our most basic test, so while the accuracy is superior to what’s offered via the Twitter API, there may be a couple of sentiment scores that need adjusting. If anyone wants the data set to examine, email me and I’ll send it off. Steve AT parnassusgroup DOT com.
Note: Thanks to the great data that Radian6 supplies, we can analyze just about any product/brand for sentiment. If you have a Radian6 account, we can analyze the data you’re receiving too. Contact us for more info.