Science Discovers the Secret to Successful Writing

[StewShearerOld]

Science Discovers the Secret to Successful Writing

Researchers from Stony Brook University have created an algorithm that can predict successful books with 84 percent accuracy.

Writing isn't always the easiest thing to do. Granted, you can say that about most any job or activity, but sometimes the art of putting pens to paper and fingers to keys does feel like a perplexing thing to get a handle on. This is especially the case when you're talking about writing something as lengthy and complex as a novel. Knowing where to start, where to go and how to get there can be incredibly frustrating; even more so when you consider that there's no way to know if anyone will even like the finished product. Expect maybe there is.

Researchers at Stony Brook University in New York have apparently developed an algorithm that can predict whether or not a book is going to be commercially successful. Utilizing statistical stylometry, a mathematical method of looking at words and grammar, the algorithm can be used to give an accurate determination as to whether or not a book destined for the bestseller's list or the bargain bin. How accurate you ask? It gets it right 84 percent of the time.

The algorithm was developed by analyzing more than 800 books, many taken from the Project Gutenberg archive of classic literature, and then comparing them to the real world success they enjoyed. Conversely, the researchers also analyzed unsuccessful books to see if they could find a common factor that led to their failure. These included both commercial failures and critical failures which they found by looking at Amazon's low selling books and by researching critical reviews.

All of this said, you're probably wondering what key elements they determined drive a novel toward success or failure. The research found that books primed to fail had a tendency to overuse verbs and adverbs and included more language describing explicit actions. Successful books meanwhile spent more time describing thought processes and also made a habit of using conjunctions more heavily. There were, of course, other factors involved -luck, for instance- which can be found in the team's <a href=http://aclweb.org/anthology/D/D13/D13-1181.pdf>official report.

Source: <a href=http://www.telegraph.co.uk/technology/news/10560533/Scientists-find-secret-to-writing-a-best-selling-novel.html>Telegraph


Permalink

 

[vid87]

Sad as that is, the thing with conjunctions explains the success of "No Country For Old Men" to me. The author really loves the word "and" - the average usage per sentence is around 3-4 and I've seen one sentence have 9.

 

[Sigmund Av Volsung]

Best-sellers aren't well written per se, just how successful movies don't have to necessarily have good cinematography(medium specifc aspects).

Also, 50 Shades of Grey was a best-seller, let's not forget that.

In addition, the metrics that define by which a book will be successful seem non-existant to me, it's all about speaking to the zeitgeist, be it criticising it(A Picture of Dorian Gray, Jane Eyre) referencing it(Cantenbury Tales, World War Z) or exploring specific aspects of it(Game of Thrones: criticism of human nature, in line with self-analysis and critique that seems to be frequent in 21st century western/northern hemisphere culture), etc.

I feel that throwing maths into the mix will be nothing more than a meta-analysis, which is unreliable.

 

[JamesBr]

snip

Note that the article isn't titled "Scientist Discovers the Secret to Good Writing". Nobody is saying that these metrics will turn you into Hemingway, it's tracking trends in popularity and taste, which pretty much defines "success". If the best writer in the world can't sell his book, he isn't successful and there are a multitude of reason why that might happen.

 

[Jumwa]

84% accuracy? I can't say I buy it. I think too many factors beyond the writing itself determine success for that kind of ratio to be realistic.

 

[FalloutJack]

I find it dubious in the extreme that someone would try to reduce that phenomenon known as creative writing to a statistic. People are unpredictable, ergo so is their response to books. What you have with the accrued knowledge here is an educated guess, but I could make such a guess too.

 

[Sigmund Av Volsung]

snip

Note that the article isn't titled "Scientist Discovers the Secret to Good Writing". Nobody is saying that these metrics will turn you into Hemingway, it's tracking trends in popularity and taste, which pretty much defines "success". If the best writer in the world can't sell his book, he isn't successful and there are a multitude of reason why that might happen.

Yes, and I said that succesful writing is also a bit trend defiant in in of itself, and is usually reflective of the zeitgeist in a way(from my experience anyhow)

 

[Falterfire]

I'm curious how they tested it. After a semester of Machine Learning I think I understand the basic theory behind it, but if they were just attempting to develop an algorithm for sorting existing data (the 800 books mentioned) it does nothing beyond describe the data you put into it.

Skimming the report, they are indeed using a Support Vector Machine on sentences taken from the books in question. Although the method is valid as a machine learning technique I'm not certain that the results are at all meaningful.

Moreover, the method they are using as a measure of success is the number of downloads the file has from Project Gutenberg. Given the nature of the books found on Project Gutenberg, I think they might skew too heavily towards just describing properties of books likely to be used in school research projects.

 

[Worgen]

Now if only we can apply it to fanfiction, the cycle will be complete.

 

[zerragonoss]

I find it dubious in the extreme that someone would try to reduce that phenomenon known as creative writing to a statistic. People are unpredictable, ergo so is their response to books. What you have with the accrued knowledge here is an educated guess, but I could make such a guess too.

Normally I would agree but here is a quote form the abstract "We examine the quantitative connection, if any, between writing style and successful literature." So they are not trying to look at the creative aspect so much as the tone and some of the technical style. Their conclusion is basically that book that are more conversational and focus more on the things people commonly care about in their descriptions sell better, not a greatly outlandish claim.

 

[FalloutJack]

I find it dubious in the extreme that someone would try to reduce that phenomenon known as creative writing to a statistic. People are unpredictable, ergo so is their response to books. What you have with the accrued knowledge here is an educated guess, but I could make such a guess too.

Normally I would agree but here is a quote form the abstract "We examine the quantitative connection, if any, between writing style and successful literature." So they are not trying to look at the creative aspect so much as the tone and some of the technical style. Their conclusion is basically that book that are more conversational and focus more on the things people commonly care about in their descriptions sell better, not a greatly outlandish claim.

But that isn't the whole picture, though. You can't render a statistic with half of the information. No wait, let me amend that, because people do that all the time. You can't render an accurate statistic with half of the information. You can determine what colors and shapes will be the most-pleasing to people, add some sound effects, and make Tetris...but it won't be a viable means of capturing art in a theorum.

 

[Hagi]

I find it dubious in the extreme that someone would try to reduce that phenomenon known as creative writing to a statistic. People are unpredictable, ergo so is their response to books. What you have with the accrued knowledge here is an educated guess, but I could make such a guess too.

Normally I would agree but here is a quote form the abstract "We examine the quantitative connection, if any, between writing style and successful literature." So they are not trying to look at the creative aspect so much as the tone and some of the technical style. Their conclusion is basically that book that are more conversational and focus more on the things people commonly care about in their descriptions sell better, not a greatly outlandish claim.

Their claim for failing books also seems fairly accurate.

If you've ever read pieces of utterly horrible fan-fiction you'll be fairly familiar with overuse of verbs, adverbs and descriptions of explicit situations.

Even if you have the most creatively wonderful storyline in mind it'll still end up being disliked by many people if you lack the writing style to describe it.

 

[Gary Thompson]

I'm dumb, somebody give me an example.

 

[Fox12]

"The research found that books primed to fail had a tendency to overuse verbs and adverbs and included more language describing explicit actions. Successful books meanwhile spent more time describing thought processes and also made a habit of using conjunctions more heavily."

So basically they're saying that well written books are more successful. Sorry scientists, it looks like you just learned something the rest of us learned years age. 5 minutes in a creative writing class could have taught you this much. Someone post a slow poke meme.

 

[BoogieManFL]

I'm dumb, somebody give me an example.

I just watched that episode of DS9 that your avatar came from yesterday.

But I can't help you with an example..

 

[Clankenbeard]

I'm dumb, somebody give me an example.

1. Another forum regular thought carefully about this request. He then joyfully typed a carefully-worded example to illustrate the described differences.
2. I read your post and posted an example.

You have to be smart and creative to write response #1. The responses say the same thing. Response #2 is still better. Nobody wants all of the useless adjectives and adverbs. Too flowerly = bad novel.

 

[Arnoxthe1]

A 16% margin of error? That's actually quite a large chunk. Not to mention the fact that as someone said, the only reason why some books are successful is because of certain factors that have nothing to do with the actual quality of the writing itself.

 

[Tiamat666]

snip

Note that the article isn't titled "Scientist Discovers the Secret to Good Writing". Nobody is saying that these metrics will turn you into Hemingway, it's tracking trends in popularity and taste, which pretty much defines "success". If the best writer in the world can't sell his book, he isn't successful and there are a multitude of reason why that might happen.

Yes. Also, opinions tend to be divided about what constitutes quality or good writing. This experiment might be more interesting and useful if they based it not on sales, but on ratings and reviews, and separated books into categories, calculating book scores for each individual category.

 

[WhiteTigerShiro]

I find it dubious in the extreme that someone would try to reduce that phenomenon known as creative writing to a statistic. People are unpredictable, ergo so is their response to books. What you have with the accrued knowledge here is an educated guess, but I could make such a guess too.

People aren't as unpredictable as you might hope.

 

[WhiteTigerShiro]

I'm dumb, somebody give me an example.

1. Another forum regular thought carefully about this request. He then joyfully typed a carefully-worded example to illustrate the described differences.
2. I read your post and posted an example.

You have to be smart and creative to write response #1. The responses say the same thing. Response #2 is still better. Nobody wants all of the useless adjectives and adverbs. Too flowerly = bad novel.

Or in short: It's in poor taste to use more words than are needed; part of the appeal to reading is letting your imagination fill-in some of the blanks.