Even with great marketing, it's tough for people to predict how well a book is received only after it's published. But that's all in the past, because there's a new algorithm that can tell if a book will be a commercial success or not before it hits the shelves. It has an 84 percent accuracy hit rate.

The algorithm was developed by researchers from Stony Brook University, New York. It is called statistical stylometry to mathematically examine words and grammar in books. The algorithm can predict better than humans if a novel will be a best seller.

The researchers used a sizable sample of classic books held in the Project Gutenberg archive and analyzed all the texts to develop the algorithm. They then compared its predictions to historical data about the success of all the books. It matched up with the real world data 84 percent of the time. Here are the key findings:

• Successful books make heavy use of conjunctions—like "and" and "but"—as well as large numbers of nouns and adjectives.
• Unsuccessful works include more verbs and adverbs, explicitly describing actions and emotions—like "wanted," "took" or "promised."
• Verbs in successful books more commonly describe thought processes—like"recognized" or "remembered."

The research is published by the Association of Computational Linguistics.

Could this algorithm have predicted what a success 50 Shades of Grey would be? That was a tough sell that one. [Association of Computational Linguistics via Telegraph Image by slightly everything under Creative Commons license]