Humour

Packaged PR, with feelings

T A Balasubramanian on the changing face of PR

“Big chiefs and big business like to indulge in media spin, and that means knowing what is being said about them. But, my dear guys, finding out is becoming ever more difficult, with thousands of news outlets, websites and blogs to monitor,” says Oza Spinoza, CEO of PR Enterprise Systems and Solutions, a company that prefers to call itself ‘The PRESS.’

The “guys” being addressed are Fernando Fizzler, Chief Marketing Manager, and Sellina Reddy, Regional Sales Manager, and their power teams.

He bows and waves a hand for dramatic effect. “As you know, baffled CIOs make the best customers. Since we have such expertise in media spin, we will launch a software package that can automatically estimate the tone of any electronic document. It can tell whether a newspaper article is reporting a CIO’s IT policy in a positive or negative light, for instance, or whether an online review of a new release of a product is praising or damning it. Welcome to the automation of PR, as designed by PRESS.”

“Well, Boss, what’s the advantage of this package?” says Fernando, ready to take notes.

“Good question, my boy. Till now, discovering whether the coverage you are getting is good or bad, negative or neutral has usually meant hiring a “reputation management” firm, and you know how much those weasels charge. They send teams of bloodhounds who will scan through everything written about a chosen organisation, person, event or issue and report back on how favourable it is.”

“Rather like a detective squad,” says Sellina, brightly.

“That’s right. As well as being expensive, this can be a long, slow process,” says Raja. “There’s a massive information overload. These news agencies are becoming news factories—a single news agency may churn out more than eight articles each hour. That’s almost 200 stories a day per news outlet. Now that’s where the opportunity for automation happens to be.”

“How did all this start, Boss?” says Fernando, busy scribbling on his pad.

“Well, previous attempts to automate this kind of analysis have used one of two techniques. In the first, called machine learning, a program is trained by being given thousands of articles already determined by a human reader to be positive or negative in tone.”

“Like training a sniffer dog to find explosives by exposing his nose to a variety of explosive chemicals,” says Sellina, who reads a lot of detective stories in her spare time.

“Exactly,” says Spinoza. “But learning in this way can lead to mistakes. For example, if a series of the training articles mentions bomb attacks on a school in Russia, the programme may incorrectly conclude that all other mentions of schools are negative too.”

“Like sniffer dogs might mistake a simple perfume spray to be a lethal chemical,” chimes Sellina.

“Then the alternative is the lexicon approach, in which certain words are classified as either positive or negative. But plenty of words can be both. For example, “the plot was unpredictable” and “the steering was unpredictable” differ by just one word. Yet the word “unpredictable” has a positive connotation in the first example and a negative meaning in the second. And even if that problem is solved, just picking up on positive or negative words can also lead to mistakes, as is demonstrated by the sentence: “Everyone told me it was terrible, that I would hate it, but in the end it wasn’t at all bad”. Now if you were to sniff out ‘terrible,’ ‘hate,’ and ‘bad’ as negative, where would you be?”

“Free and happy,” says Sellina.

“I mean, do you realise the difficulties of interpreting words in different contexts? So what we have come up with is a program called PRUNE, short for PR’s Universal New Extractor, which uses algorithms to tease out grammatical components, such as nouns, verbs and adjectives, and identify the subjects and objects of verbs. It can even analyse pronouns like “it”, “he” and “her” to work out what words or concepts they are referring to. However, what makes this kind of analysis so challenging is that key words in a text often offer no clues as to what sentiment they carry. For example, if I say ‘Why should I bother going to the movie?’ it may seem to be neutral to a software program, but you know that I am merely rhetorical and negative about going to the movie. So some of the toughest riddles in comprehension, such as identifying irony and rhetoric, are likely to remain unsolved for some time.”

“How is our product an improvement over the earlier packages?”

“PRUNE is emotionally sensitive. Having an understanding of grammatical structure makes it possible to filter out words that are not relevant to the feeling of the article,” Spinoza says. “So instead of assuming certain words, such as ‘unpredictable’ or ‘rubbish’, are positive or negative it allows the structural context to disambiguate them.”

“What is disambiguate, Boss?” says Fernando, frowning.

“The opposite of ambiguate,” says Sellina, smirking. “Which means creating confusion when things have more than one meaning.”

“That’s right,” says Spinoza. “We’re in the business of disambiguation, or spin, and it is now our privilege to offer PRUNE to the world, changing the face of PR forever.”

“Mr Spinoza, is this software infallible? I mean, does it work in any situation, however sticky?” says Sellina.

“It doesn’t get it right all the time,” Spinoza admits, “but then neither do people. Three expert readers are likely to agree about an article 85 percent of the time, and about 90 percent of non-experts will agree with this consensus. Now PRUNE agrees with the same expert consensus about 80 percent of the time, which is pretty good, considering that it almost equals the experts.”

“Great, Boss,” says Fernando. “So now we don’t need those reputation management bloodhounds any more to tell us how we’re doing?”

“Not exactly, Fizzler, though I see it might be possible to get CIOs educated one day to the same level as the bloodhounds. PRUNE will not take the humans out of the equation, because someone is still going to have to evaluate the software’s report on each article. But since the program will list items in terms of how positive, negative or neutral they are it is possible to zero in on the most relevant items.”

“Now that will allow us to prioritise, and do the job much faster, Boss,” says Sellina.

“Right on the button, Ms Reddy. Speed is the name of the game. While you might be able to scan 10 articles an hour, PRUNE here can zip through 10 a second. Now that’s what I want you guys to go out and tell the world, pronto. So pack up your Powerpoints, put on the war-paint and go to market. As of today, we are getting PRUNE on the road to make PRESS the last word in PR.”