()
It was a tossup between a game-theoretic examination of the toilet seat problem and the UnSuggester, which takes the title of a book you like and tells you what books to avoid. I gave it "Applied Cryptography" and it told me to stay away from "The Devil Wears Prada," whcih is something I had kind of planned to do, anyway.
http://www.librarything.com/unsuggester

Melinda Shore - Software longa, hardware brevis - (Email Removed)

Prouder than ever to be a member of the reality-based community
1 2 3 4 5 6 7 8
I gave it "Applied Cryptography" and it told me to stay away from "The Devil Wears Prada," whcih is something I had kind of planned to do, anyway. http://www.librarything.com/unsuggester

Okay, that's weird. I plugged in "And the Saw the Angel" and it spit out "The Devil Wears Prada" as the number one unsuggestion.

Shelly (Warning: see label for details)
http://www.cat-sidh.net (the Mother Ship)
http://esther.cat-sidh.net (Letters to Esther)
http://www.librarything.com/unsuggester

Also, I'm disappointed and saddened that there are so few entries for "The Diary of Anne Frank," which was my first attempt.

Shelly (Warning: see label for details)
http://www.cat-sidh.net (the Mother Ship)
http://esther.cat-sidh.net (Letters to Esther)
Also, I'm disappointed and saddened that there are so few entries for "The Diary of Anne Frank," which was my first attempt.

Well, we aren't exactly a nation of readers. I read a bunch of economics blogs and there's a libertarian one I really like even though I'm not a libertarian, because the author is an extremely literate, clever, enjoyable nerd, and his positions have an unusually high degree of internal consistency.
So in the process of wondering about how many people in an airport are actually local residents he accidentally said some negative things about Minneapolis even though he actually likes it and in a follow-up post he laid out some great things about Minnesota in general, and in the comments section someone mentioned Vince Flynn as a Minnesotan he (the commenter) thought was particularly great. Now, I don't know if you've bothered to read Vince Flynn, but aside from his use of right-wing political assassins as American heroes he's basically completely illiterate and I have to assume that the commenter singled out Vince Flynn either because he likes the plots or because he's most comfortable with monosyllabic words.

Melinda Shore - Software longa, hardware brevis - (Email Removed)

Prouder than ever to be a member of the reality-based community
http://www.librarything.com/unsuggester

Okay, so I tried it. Put in 'Wide Sargasso Sea' (by Rhys) and my #1 unsuggestion is
'Desiring God: Meditations of a Christion Hedonist', which actually sounds interesting. In fact, most of the top works are somehow theologically related. What the hell?
Then I tried the BookSuggester, and tried the same input. Here is what I got out.
To the Lighthouse (I haven't read this one, but generally don't like anything by Virginia Woolf)
The Mill on the Floss (Hated It)
Mrs. Dalloway (Hated It)
The Portrait of a Lady (Hated It)
Sons and Lovers (Didn't hate it, didn't particularly like it either)

The funny thing is that I have read and either intensely disliked or was meh about 13 of 16. Of the remaining 3, the only one I am even remotely interested in reading is The Remains of the Day. I have read other things by Ishiguro and liked it.
Suja
'Desiring God: Meditations of a Christion Hedonist', which actually sounds interesting. In fact, most of the top works are somehow theologically related. What the hell?

"Network" stuff is very hot right now, and they're analysing the individual data sets in their databases and clustering stuff that tends to appear in the same data sets. That means that you're pretty much at the mercy of the tastes of the people who choose to enter their libraries into that website, a self-selecting group that could very well have some unattractive reading habits (says Melinda, who has some unattractive reading habits). It might be revealing to do the same kind of network analysis of the people who uploaded their libraries - for all we know, the site was announced in a church newsletter or something and the newsletter's readers are overrepresented in the population of site users.

On the other hand it just might be a clue as to why the "War on Terror" is going the way it is, since there's heavy use of the same kind of analysis to identify terrorists.
Melinda Shore - Software longa, hardware brevis - (Email Removed)

Prouder than ever to be a member of the reality-based community
Well, we aren't exactly a nation of readers.

Well, no, we aren't. But Anne Frank? It is for to weep!

I once had to go to a pretty pointless seminar on "What's wrong with management?" The only thing I took away from it, and that was because it was amusing under my then-current working circumstances, was that "A fish rots from the head down." But, hey, I hear that the whiny smirker has actually held a copy of Camus' "The Stranger" in his very hands, so maybe he'll become less rotten by osmosis?
So in the process of wondering about how many people in an airport are actually local residents he accidentally said ... a Minnesotan he (the commenter) thought was particularly great. Now, I don't know if you've bothered to read Vince Flynn,

Never heard of him. After Googling and finding out he's a consultant for "24," I think it's a safe bet that I'm better off remaining ignorant of his, um, oeuvre.
but aside from his use of right-wing political assassins as American heroes he's basically completely illiterate and I have to assume that the commenter singled out Vince Flynn either because he likes the plots or because he's most comfortable with monosyllabic words.

The two aren't mutually exclusive. In fact, I'm betting they tend to go hand-in-hand.

Shelly (Warning: see label for details)
http://www.cat-sidh.net (the Mother Ship)
http://esther.cat-sidh.net (Letters to Esther)
"Network" stuff is very hot right now, and they're analysing the individual data sets in their databases and clustering stuff that tends to appear in the same data sets.

Having been there and done that, I must say that there is something very off in their clustering algorithm. Or maybe it's how they derive the 'Un' part that's off.
I tried again with The Alienist, and got back VERY similar stuff, although near as I can tell, there is virtually no similarity between the two books I used as input (unless you count how one is the story of the mad woman in the attic, and the other is about a serial killer, who couldn't possibly be altogether right in his head). This time though, they threw in a couple of books on programming and compiler design for good measure.
some unattractive reading habits (says Melinda, who has some unattractive reading habits).

Ditto. I haven't had time to read anything that can be considered 'deep'. Hell, these days, I'm lucky to get around to the Comic pages in The Post.

Suja
Having been there and done that, I must say that there is something very off in their clustering algorithm. Or maybe it's how they derive the 'Un' part that's off.

Well, back when I was doing graduate work in information retrieval, clustering (although not by network analysis) and vector-space (measuring angles between term vectors) techniques were two of the things we were looking at. And the ugly truth is that pretty much any of these analytical approaches underperformed just taking a list of terms in a document, stripping off prefixes/suffixes, ordering by term frequency, and dropping the 25% most frequently occurring terms and 25% least frequently occurring - that is to say, building an index from the words that appear in the document. It could very well be that network analysis gives crappy results, too.

Melinda Shore - Software longa, hardware brevis - (Email Removed)

Prouder than ever to be a member of the reality-based community
Show more