Cornell University
You have good intended, the truth? Which is a fake review and that is real?
By Rob Lovitt, msnbc.com contributorWhen it comes to online hotel reviews have the wisdom of the crowd and are solitary Wolf liars who plant a bogus reviews.
You think you can spot the difference? See the above two reviews and may find that your BS detector is not adjusted as you think.
According to research conducted at Cornell, the usual man is able to correctly identify the fake reviews about 50 percent of the time, while software developed by the team of the research is almost 90 percent accurate.
To test his theory, researchers compiled 400 truthful reviews and 400 made up such covering 20 Chicago hotels and develop an algorithm that analyzes them based on the parameters, ranging from experiences cited punctuation and parts of speech.
Truthful reviews tend to focus on the spatial data, "said lead researcher Myle Ott and use more nouns, verbs and adjectives. Top indicators include words such as "location", "bath" and "useful". Real reviews also uses more dashes, ellipses, and parentheses.
Deceptive ones, on the other hand, focuses on the external aspects and details. Top indicators include words such as "I," "my husband" and "holiday".
What it all means? For one thing it must provide some clues as to which of the above survey is misleading. (Answer: this right.) Move forward, it may also set the stage for improving the reliability and happy hotel stay.
"Ultimately, intuition might not be the best thing to rely on," says Ott, "but there is hope for such computer programs which make distinctions for us."
More stories that maybe you want:
Rob Lovitt is a long-time travel writer, who still believes the journey is as important as the destination. Follow him on Twitter.
0 komentar:
Posting Komentar