1. This is a still from the end of yesterday’s Conservative video, which is just 1 min 20 secs. I thought it was quite an effective compilation of short clips, including Harriet Harman and Ed Balls saying that taxes and debt would be higher, and Ed Miliband saying he was opposed to an in-out EU referendum.
The Tory clips were all right except when Iain Duncan Smith appeared.
Labour’s effort the day before was also slick, although twice as long. The patronising “ordinary people” had been rebranded as “everyday families”, which is a change without a difference.
West Upper Plastic commented on the “very clever use of Ed Miliband in that video” (he doesn't appear).
2. Polling news. YouGov for Times Red Box yesterday: only 3 per cent could recall any part of Green Party leader’s name, including “Natalie someone Australian” and “Nicola Burnett”. So much for the Green surge. But I did come across an old article before the last election that claimed Samantha Cameron voted Green as a student in 1992. I wonder if it is true.
3. Meanwhile, a TNS poll of Scotland found the SNP has a 16-point lead over Labour, pretty much the same as other polls, and an awful lot of Labour seats gone. And a new Survation poll of Thanet South puts Nigel Farage in the lead, with Labour in second place. I now think UKIP will win five seats: Clacton, Rochester, Thanet South, Great Grimsby and one surprise.
4. A footnote on those immigration figures from yesterday. Non-EU immigration – the bit Government can control – is still higher than EU immigration, and increasing faster. So it is not simply that David Cameron made a foolish promise on which he had no power to deliver while Britain remains a member of the EU.
5. I came across a robot sarcasm detector last week, which could be useful. TheySay, a spin-off from Oxford University, has started using “computational linguistics” to assess the mood towards the four party leaders on Twitter and blogs. This is its scoreboard for the 24 hours until this morning:
I spoke to Karo Moilanen at the company about how “sentiment analysis” works. He told me the algorithm detects positive and negative sentiments associated with the leaders, and can even recognise a double negative as a positive, for example, “kill bacteria”.
What about sarcasm, I asked, thinking about how Twitter works. “We have a rudimentary sarcasm detector,” he said. “There are patterns which tend to correlate with sarcasm.” But how accurate is it? “Sarcasm is hard for people to detect. Human accuracy can be as low as 40 per cent.”
TheySay “trains” its computer programme by feeding it texts that humans have marked as being sarcastic. “Algorithms can hence learn that sarcasm tends to involve cases in which someone likes something negative,” said Moilanen, “or conflicting or abrupt changes of sentiment between strongly positive and negative words and phrases.” He said that computer algorithms can detect sarcasm between 55 and 95 per cent of the time, depending on the study, with an average of 77 per cent.
People say that irony doesn’t work on Twitter. Now we know: it works 77 per cent of the time.
6. And finally, thanks to Moose Allain for this:
“akld;mfpkowmmv.v,, doigj oij,af,.gmjs[oimgij[oic jocjkqdgfaksdmg[ain.c,f-o”
Extract from An Infinite Number of Monkeys: the Early Years.Reuse content