BookVibe mines the Web to find your next read

[Alexander Turcic]

There are many good books out there. The only problem is finding them. You can follow recommendations from Amazon and roam social platforms such as MobileRead or Goodread. Or you could trust the proprietary artificial intelligence of BookVibe, a new book discovery engine that extracts data from conversations in social media, including Twitter, to identify what's currently hot and what's not among users.

Vindu Goel of the NY Times gave BookVibe a spin:

The service is still in beta, and it shows signs of being a work in progress. Extracting real meaning from the shorthand found in 140-word tweets can be a challenge for humans, let alone computers. Some books popped up on the recommended list because their author had mentioned them. Some reviews were missed because the Twitter user offered a link to an external review without summarizing it in the tweet.

But over all, I found BookVibe to a valuable single-purpose tool and an indication of what’s possible as social media search technology becomes more sophisticated.

Full article @ NY Times Blogs

[via dbw]

[medard]

That looked nice and useful at first sight. But reading the NYT article gave another impression to me:

In addition to the free consumer-oriented BookVibe, which will soon come out in a Facebook version, the company sells a more in-depth set of tools to book publishers, authors and retailers to help them understand how books are selling and where and what people are saying about them.

So this is just another 'Big Data' recycling bin. Personally I prefer the advice of experts rather than emotions of the masses or recommendations from the robots. At least for now.

[Alexander Turcic]

Originally Posted by medard

So this is just another 'Big Data' recycling bin. Personally I prefer the advice of experts rather than emotions of the masses or recommendations from the robots. At least for now.

True, who knows, perhaps if the sample size is big enough, their recommendation will ultimately reflect Amazon's top bestsellers lists. Nothing much gained then.

[Graham]

Routines that mine my own reading preferences for recommendations I can understand. Routines that mine everyone else's are only marginally useful.

Graham

[Ravensknight]

Originally Posted by Graham

Routines that mine my own reading preferences for recommendations I can understand. Routines that mine everyone else's are only marginally useful.

Graham

Yep. I don't care what others have read/are reading. My tastes don't coincide with theirs a lot of the time. A program that doesn't base its conclusions on my data doesn't seem that helpful...

[tbar98]

Hi Everyone - thanks for the great discussion and comments regarding Bookvibe. We did want to clear up one misconception regarding the chain - the advanced algorithm used doesn't just randomly search and present the results based on what others are saying. Rather, it looks at a person's 'affinity' including Twitter profile, Tweets and Retweets, the makeup of followers, etc., and matches that information against the Tweets to get targeted recommendations. So for instance if I say I am a Tech enthusiast in my profile, the algorithm would know to pick up a book recommendation on 'mobile gaming' that someone I follow recommended.

We love the comments and will continue to keep listening and honing the technology based on the feedback to make Bookvibe the best book recommendation engine possible.

[speakingtohe]

I don't think that tweets and twitters are of much use in determining the overall merit of a book.

It tells you that someone or many someones liked the book, or the author and his relatives stay up all night and tweet.

I can tell what is popular generally by reading the various bestseller lists or going to a drugstore.

I want to find books I would enjoy possibly based on books I have enjoyed or based on genre or somebody stating a reason that they enjoyed it. I read books in most genres, and a sincere I just like these books will often make me read one. To me tweeters and twitterers are much like the names imply, tweeters and twitterers. They have a right to be heard perhaps, but actually listened to by me is not likely to happen unless there is a blackout for my best friends football game and he wants a play by play

Helen

[JD Gumby]

Originally Posted by speakingtohe

I don't think that tweets and twitters are of much use in determining the overall merit of a book.

It tells you that someone or many someones liked the book, or the author and his relatives stay up all night and tweet.

From the sounds of it, it moniters *YOUR* feed, not Twitter as a whole. Well, at least the user version, anyways. (though it's fairly obvious that it's just a data harvesting tool so they can to sell to publishers)

[DiapDealer]

I'm intrigued enough to at least think about checking it out. But the website only has a place for me to enter my email address to "get a weekly email of book recommendations" from my friends. If a weekly email is the extent of the service, I'm not inclined to "sign up" for another online account. If that's NOT the extent of the service, it would behoove them to indicate exactly what I'm "getting" in exchange for giving out my email address (preferably before I do so).

You have to admit, in this day and age of sparkly apps and the ability to sort data in any way a user might want to see it... a weekly email isn't all that enticing.

[Graham]

What other methods does it use? I don't use Twitter, so would I get no recommendations?

Graham

[jineapple]

To be honest, I don't see how this is supposed to work better than just using a list of books you've read and rated like e.g. goodreads does (and tbh even the goodreads algorithm is pretty poor, I still get a lot of books I'm not interested at all in my recommendations)