A new tool launched today that promises to delve into the sentiments being expressed about particular topics right across the Social Web.
Veooz (pronounced, we are informed, ‘views’) tracks trending topics across the major platforms and uses semantic algorithms to determine how positive or negative the general sentiment is. This is a step forward from simply knowing that a topic is trending. Analysis of this kind will help businesses avoid finding themselves in the midst of a twit-storm for leaping-without-looking into online conversations.
Let’s have a look
On the front page of veooz.com, there are a number of trending topics to have a look at. For instance, after Prime Minister’s Question Time this lunchtime, it is not surprising that ‘David Cameron’ is being widely mentioned. Here’s a snapshot of Veeoz’ interpretation:
It’s just semantics
It’s not surprising that a Prime Minister who is not wildly popular (even within his own party) should have a negative rating, perhaps. But have a close look at the Influential Tweets. The top one is distinctly negative. What about the other two? @SaveTheChildren use the word ‘pressure’, but given they are a charity and a pressure group hawking a petition, this can hardly be seen as negative. The bottom tweet is telling, though. The key words here in rating sentiment are likely to be “condemnation” and “ugly”, however the PM joining in with the condemnation of condemnable scenes is probably a good thing.
Let’s take this further
Danny Rose is a young British footballer who was subjected to racist abuse last night during the game in Serbia. He got himself sent off for his reaction during a skirmish after the match. Many would think his reaction, although hot-headed, was quite understandable; perhaps even laudable. Why then such a hugely negative rating for the young man?
Look again at the ‘Influential Tweets’. @piersmorgan uses the words “Disgusting” and “disgrace”, but not about Danny. Instead they are directed at the scenario he found himself in. And yet, “Danny Rose” has a 96% negative rating.
Clearly it is early days for sentiment analysis and semantic search techniques. Pulling out the level of detail that would be required to provide an accurate assessment was the kind of thing Alan Turing dreamed of. We’re obviously a long way away from that. Especially for a free-to-use system that is still in beta.
For big brands, though, with large and vocal followings there may be a degree of insight available to those with the time and energy to mine the data. Amazon will be rightly pleased with the reception Kindle Paper White appears to be getting, for instance.
For you and I, though, let’s remain sceptically intrigued.