Statistically Improbable Phrases is an Amazon feature launched in 2005 to identify a book's most unique word combinations compared to other books indexed by Search Inside's program. SIPs are one more potential way for readers to find relevant books, especially niche content.
For books enrolled in Search Inside, you'll find SIPs displayed on the book's detail page under the heading #Inside This Book#. For example, the book The Da Vinci Code has these SIPs:
cilice belt
seeded womb
pope interred
lame saint
lettered dials
tracking dot
inlaid rose
corporal mortification
rosewood box
sacred feminine
depository bank
royal bloodline
stone cylinder
sweater pocket
Amazon's SIPs are hyperlinked, so clicking on them brings up a display of other books containing the same SIPs. From here, users can click through to see where the SIP appears in each book, using the Amazon Online Reader.
It's surprising to see how seldom certain word combinations occur among millions of books. For example, the first SIP given for The Da Vinci Code, "cilice belt," appears in only one other book--a book about _ The Da Vinci Code_.
Amazon is considering ways to merge SIPs with its recommendation system, or use SIPs for new kinds of services, like answering questions using authoritative texts.
Along with SIPs, Amazon displays other text statistics, including a list of capitalized phrases and a list of the 100 most frequently appearing words in the book, called the concordance.
How helpful these extra features are for shoppers isn't yet clear. "When we expose new features, we measure how they change the customer's behavior," says Amazon's Bezos. "For example, does it take the customer fewer steps to find what he or she needs? This is hard, because you are measuring human behavior. There are some things that customers are delighted about immediately, and there are other things that they have to get used to."