N-gram Analysis of the New York Times Weddings Section

http://news.rapgenius.com/Atodd-when-harvard-met-sally-n-gram-analysis-of-the-new-york-times-weddings-section-lyrics

by lil_tee • 13 years ago

250 64 13 years ago

news.rapgenius.com

bigtech• 13 years ago

Are all wedding announcements posted, or does The Times have an editorial role in which announcements are printed?

daniper2• 13 years ago

They have editorial discretion. I'd be curious to know what the "acceptance rate", so to speak is... I know for a fact that a couple I know was rejected specifically because one of them was undergoing gender reassignment and the NYT didn't know which personal (gender) pronoun to use for him.

mason240• 13 years ago

The NYT said they rejected it for that reason, or that is what you friend assumed to be the case?

mtdewcmu5• 13 years ago

Maybe a different section would have printed it. I'm kind of interested.

Haul4ss• 13 years ago

AP StyleBook says to use the gender the individual identifies as. Thus why all the news media now reports on Chelsea Manning.

mturmon2• 13 years ago

The Times certainly picks and chooses which are printed. That's the whole reason for the interest in the topic. It's a perfect collision between young-adult ambition and old-school establishment vetting.

I understand that, like college admissions, you can hire a wedding planner or consultant who can considerably raise the chances of your wedding being listed.

The NYT obits are another interesting read.

WillyF3• 13 years ago

We used a high-end wedding planner, and when we asked her what to do for the NY Times announcement, she told us to submit via the form on the website. Maybe she pulled some strings behind the scenes, but it seemed to us that she had no pull whatsoever on that front.

MichaelGG2• 13 years ago

What is the point of this? Is it some kind of odd "old-school" status thing? The upper class equivalent of being on the 8 o'clock news for a 1 minute interview? Although perhaps this is only for people already investing a huge amount of money into the wedding ceremony?

mtdewcmu5• 13 years ago

Considering that there are 8 million people living in New York, including everyone would present some serious challenges indeed.

WillyF3• 13 years ago

I don't have any stats, but I know that very few of those who submit are selected. My wife and I received a long form write-up (just under 700 words). The process started with our submission on the website, but then we got a call from a writer a few weeks later. Between talking to us, our parents, and our minister, I'd say he put at least 10 hours into researching and writing.

A day or two before the wedding, he told us that he wrote both a long form piece and a shorter, more typical piece. He wasn't sure which would get published, but he was obviously pushing for the longer piece to get in. It did.

Oddly enough, our write-up isn't included in the Rap Genius dataset. Maybe it's too recent or the longer write-ups aren't included.

daniper2• 13 years ago

The data-set includes only the announcements, not the stories published under the "Vows" column.

MichaelGG2• 13 years ago

Not to be rude or offend you, but, why? You're saying a newspaper spend 10 hours researching and writing a 700-word article about your wedding? Is this a "local flavour" type article, like they might do about assorted residents? Or are your families famous or ? Did you pay for this kind of placement? Perhaps I'm socially inept here, I just feel like I'm missing something.

As far as being in the dataset, certainly they aren't analyzing every flavour-style thing article, just "announcements"? Just like a 2-page life-in-review article on someone famous when they die doesn't really go in the obit section, does it?

jmduke• 13 years ago

Given how horizontally expansive RapGenius is trying to be (this is the first time I've seen NewsGenius, but I'm familiar with PoetryGenius et al -- there's an annotated Iliad that's pretty cool), I'm wondering if they'd be better off as a layer or a plugin as opposed to a stand-alone site. I'm much less tempted to visit the site for each individual story that pops up than I would be to peruse the annotations as I browse normally.

Either way -- stuff like this is a delight to read.

steveklabnik2• 13 years ago

My understanding was that RapGenius was always a text annotation platform, with the hip-hop stuff serving as an exemplar, rather than being the actual product.

jere• 13 years ago

Always a text annotation platform? That's a bit revisionist from what I've read.

Fred Wilson initially told them "I think lyrics is a very crowded space and almost entirely reliant on Google for traffic" and they admitted "our pitch back then was a bit too lyrics-focused.."

http://news.rapgenius.com/Lemon-how-rap-genius-raised-s18m-i...

You can find those quotes in an annotation in the above link (which makes me realize the problem with annotations is you can't ctrl-f them).

why-el• 13 years ago

A layer is not a good idea from a business point of view in my opinion, given how most websites (Facebook, Twitter with their API changes, Spotify with their new messaging system and so on) are trying to lock you in. Maybe simply build a way to take you back directly to RapGenius, like a annotate-this-on-rg button.

tlrobinson• 13 years ago

Most users have no idea what a plugin is or how to use it.

That said, it might be nice to have an optional plugin to easily discover annotations and add your own to pages as you browse.

qwerty_asdf• 13 years ago

Funny you should mention that.

Given that that's the defense people seem to proclaim every time someone mentions that disabling JavaScript is now buried as an arcane config flag in Firefox.

They say: "just use a plug-in", "just use an add-on"...

What they mean to say is: "just don't disable javascript at all... ever."

jsnk• 13 years ago

Is it just me or does anyone else dislike reading articles like this with dark background and light text?

fatjokes• 13 years ago

Personally I prefer it. Dark text on a white background (aka, the norm) often requires me to turn down the brightness on my laptop.

sengstrom• 13 years ago

"This makes it possible to rigorously test our intuitions about trends like."

let me fix that for you

"This makes it possible to put numbers on our preconceived notions and play around with them."

It may be entertaining, but rigorous? I don't think so.

mjn• 13 years ago

Yes, generally n-gram-based analyses are a huge minefield. Computational linguists do use them, with a lot of caveats and careful analysis of confounding factors.

One simple one that comes to mind here is that you need to analyze to what extent changes over the period of the data set are caused by underlying societal changes, versus changes in the NYT itself; the end result will be a mixture of those two changes, some of which may be magnifying and others offsetting. The 1980 NYT and the 2013 NYT are not the same newspaper, not edited by the same people, not sold to the same readership demographics, and not soliciting the same advertisers, so it's somewhat questionable to treat it as a stable proxy for a social group.

Another common pitfall is language change screwing up all kinds of measures (since n-gram models just work on word counts). For example, if two words are used roughly interchangeably in 1980, but by 1990 one of them has fallen out of usage, and been replaced wholly by the other one, searches for just the one word will look like the word's on an upwards trend, but it would be misleading to infer an increase in the underlying concept over the period. Of course, you can account for this by merging words into equivalence classes (most analyses will do basic stemming and merging of alternate spellings), but you have to be very careful to get all the equivalence classes (which is not a well-defined notion). Just a list of the top words in a year will tend to be some mixture of 1) top concepts; and 2) concepts expressed using only a small number of wording variations, so their count doesn't get diluted.

beachstartup• 13 years ago

yeah the guesses on ethnicity based on common last names? "chang chen wong" ... rigorous indeed.

wonnage• 13 years ago

Not to take an entertaining post too seriously, but when your graph scale ranges from 0 - 0.02% the statistical significance is dubious.

lbarrow• 13 years ago

Thats... not how statistical significance works.

rm9992• 13 years ago

Sure it is. 60,000 weddings * 0.02% is an expected number of 12 positive examples, which really isn't much. Assuming a binomial process, n=60000 and p=0.0002 gives a 95% confidence interval of 5.2 to 18.6, which is a really wide range when you want to show trends. I don't know if the percentages are by year, but if they are the issue is even worse.

The post just does a good job of hiding it by smoothing the plots. Compare an unsmoothed plot: http://www.weddingcrunchers.com/?q=Democrat%20%2B%20Democrat... with the smoothed plot in the article: http://s3.amazonaws.com/rapgenius/HhvuocYI3raAnYpWPE4HaeCh9a...

While the % of republicans does appear to fall, the % of democrats in the last year is lower than in the first year, the opposite of the conclusion they want you to draw!

davvolun• 13 years ago

As you said, caveat 'entertaining' blah, blah. That said...

That's actually a slightly dubious analysis. The question you need to ask is 0.02% of what? In this case, I would take a guess it means 0.02% of all the words analyzed. As a very simple example, imagine analyzing all the letter in a book. If English were perfectly balanced, we expect to see all 26 letters at 1/26 or 0.038%, so seeing the letter 'e' appear at say, 1.0% (or even 0.1%) would be a notable statistical result.

edu• 13 years ago

I'd say it's 0.02% of all the weddings, at least the axis is "NYT Wedding Frequency" not "Word Frequency".

mtdewcmu5• 13 years ago

If all statistics were required to be rigorous, there'd be a whole lot less to read about in the newspaper, that's for sure.

misiti37802• 13 years ago

Really interesting - how did you guys downloaded the 60K articles (is there an API i do not know about)? Also - what graphing lib are you using (I see it is not d3)?

wwarnerandrew2• 13 years ago

The graphing library is highcharts (http://www.highcharts.com/)

misiti37802• 13 years ago

Does anyone have any clue how they would download these 60K articles ???

bradleyjg• 13 years ago

A 1987-2007 dataset is available from UPenn:

http://ldc.upenn.edu/Catalog/CatalogEntry.jsp?catalogId=LDC2...

jlgreco• 13 years ago

I am very surprised by the prevalence of "was graduated from". I have only rarely heard that in 'real life', is the NYT's style guide enforcing this usage?

jcampbell1• 13 years ago

Those are the most popular phrases in 1980. The times no longer uses "was".

lifeisstillgood2• 13 years ago

I expect so - graduation is something that happens to you

Casseres• 13 years ago

Agreed. When people say, "I graduated college", they are saying the college graduated from them, not they graduated from college. The correct usage should be, "I graduated from college."

the_watcher• 13 years ago

Was this designed specifically for Katie Baker (she writes summaries of the NYT Wedding Section for Grantland)?

Alex3917• 13 years ago

N-gram Analysis of the New York Times Weddings Section

Related Stories