Showing posts with label wall street journal. Show all posts
Showing posts with label wall street journal. Show all posts

Thursday, August 21, 2008

Reporting on the reporting of Vytorin Cancer Data

Our colleagues at The Wall Street Journal used some data from Dow Jones Insight to blog about the size of the spike of press coverage of a possible link between use of cholesterol-fighter Vytorin and increased risk of cancer.

A few of the comments were hostile toward the post. This one was from someone called "Steve Walker":

So now the WSJ is going generate and report statistics on how frenzied the press becomes over junk science flowing from questionable meta-analyses? Then try to analyze why some junk science leads to less or more press frenzy?...

I disagree with this criticism. I think reporting on the media coverage of a story such as this is valid as is the methodology used to count the media mentions.

Wednesday, March 05, 2008

2000 Press Mentions of 'Stagflation': NPR

There was a "cough-and-you-miss-it" mention of Factiva on NPR's Morning Edition Tuesday when Wall Street Journal economics editor David Wessel said he found 2,000 mentions of "stagflation" in the Factiva database. (Thanks to attentive listener John C. in New Jersey. )

We know that journalists use Dow Jones Factiva's archive of mainstream media to do this kind of light text mining all the time.

I took David's statement a bit further and looked for the word in the past 14 months using Dow Jones Insight. I'm not sure what time frame he used. I found 4,603 documents mentioning that term during that time and certainly a lot more interest of late. This was from about 6,000 world-wide mainstream media publications and wires.




(Full disclosure: The Journal and Factiva are both owned by Dow Jones, which also pays my salary.)

Wednesday, June 06, 2007

Editorial Integrity Questions - Redux

Journalists, politicians and others concerned about the quality of the newspaper, which is regarded as one of the best in the world, are expressing fears that it ... will change editorially if they are added to Mr. Murdoch's substantial international press empire.


A quote from another article this week about Rupert Murdoch trying to buy The Wall Street Journal?
Nope, that was from The New York Times on Jan. 21, 1981, right before Mr. Murdoch purchased the august Times of London.

Monday, May 21, 2007

'New Media' Makes a bit of a Comebank

Mining the term "new media" from The Wall Street Journal over the past seven years shows that it was mentioned more frequently in 2006 than at any point since the end of the dotcom era. Note the decrease in the term "dotcom" over the same period.

Wednesday, February 28, 2007

Hillary gets more mentions than Anna Nicole - barely

Do you think Hillary Clinton's people see it as a victory that they edged out Anna Nicole Smith or a defeat that they just barely edged her out?

I just don't get tired of fun with text mining.
(Note: This is a continuing feature of Factiva Insight's text mining results being run in The Wall Street Journal.)

Friday, February 23, 2007

Spell Checking the Media

This is another example of a fun use of text mining -- run by The Wall Street Journal sourced from Factiva Insight

Wednesday, January 17, 2007

Top Cliches of 2006 - from Factiva Insight


The Wall Street Journal ran what's become known around here as the "Factiva Cliche Index" today. (And lest you're wondering, just because Dow Jones now fully owns both brands doesn't mean we Factiva folk can walk right over to the editors and say "run this." Not that it hurts to be in the same building, or anything.)
Bottom line, this is a great example of text mining put to use to give information that single searches can't do. Clichés might be a fun example, but look for more hard-hitting ones coming soon.