Software Tracks Blogosphere Mood Swings

holy_calamity writes "Dutch researchers have figured out a way to measure the mood swings of the blogosphere. It can pick up peaks of flirtiness from bloggers around Valentine's Day and drunkenness at weekends, the plan is to create a search engine that returns the prevailing mood in the blogosphere about a topic. Companies are already interested in using it to track consumer confidence. What's the mood of Slashdot on this one?"

Comments

Been there, done that.

[XorNand]

This is old news. Just blogs? Bah! Color me unimpressed. I've already harnessed the power of the Internet to track the mood of the entire planet: http://www.howisyourday.com/

Re:Been there, done that.

[Umbral Blot]

Are you trying to be funny? Getting a small group of people who speak english to and are willing to put up with voting / verification is in no way a representitive sample. Now if you went into their blog and determined their mood from the text, that might be cool.

Re:Been there, done that.

[JordanL]

FTA: On Valentine's Day, there is spike in the numbers of bloggers who use the labels "loved" or "flirty", but also an increase in the number who report feeling "lonely".

So let me get this straight... all this to discover that people either think about their relationships or lack or relationships on Valentines Day...

This is the ultimate geek software. "What is this strange thing... emotion... I'll build a software program to explain it to me."

Re:Been there, done that.

[Nesetril]

Not to mention, TFA doesn't explain how this is accomplished from a technological point of view. Makes me suspect that all they are doing is just a frequency analysis painted with Kinkaid colors and various insane words from the "blogospheric" vocabulary.

Re:Been there, done that.

[daviddennis]

Actually, the article was pretty clear on how it worked, but I'll explain it a little better since I guess (from your post) it was confusing.

When you write a blog entry in LiveJournal, you're give an opportunity to select a "mood" from a dropdown list of moods. So you can say you're

happy
sad
loved
excited
lonely
(etc)

by just picking the appropriate word.

Now, as you know, emotional data taken from a dropdown list at the end of writing a blog post might not be worth taking all that seriously, but it's data, and you can try to analyze it.

Now, some people are laughing at this by saying that it should of course be obvious that people are likely to be feeling loved or lonely around Valentines' day. But actually this is an important observation, since it says that what people pick in the dropdown can be related to real events. Of course we know people are loved/lonely on Valentines' day; what we didn't know is if what they picked on the dropdown was meaningful. Now we know it is, and so (in theory) we can use this to predict events or people's behaviour based on what they say.

The Harry Potter example showed that this could in fact be done, and this means that further reasearch might be promising. For example, let's say there was a "suicidal" mood in the list. It would be interesting to track whether actual suciides were predictable or at least more likely after showing those moods, so that an early warning system for such behavior could be created.

On paper, it seems possible that lives could be saved that way, which makes this a non-trivial application indeed. To support my theory, note the previous news reports we've seen here that note that suicidal behavior was often predictable in hindsight from what people wrote on their Myspace profiles. If this could be determined from moods, which are trivial to check automatically, it might be a very interesting result indeed.

Hope that helped people's understanding.

D

Re:Been there, done that.

[bnf]

So if there is a corelation between mood and blog posts how about using this data set to "prime the pump" of a Bayesian filter which would analyse blog posts on other sites and determine how likely it is that the blog post is one of these moods.

Re:Been there, done that.

[gold23]

Don't be too hard on yourself. I got it. I think it was the "Bah" that really tipped your hand.

Blogosphere Mood

[NitsujTPU]

My mood always turns sour when people refer to the "Blogosphere."

I'll take a few fewer buzzwords a day, and call my Dr. next week to see if the situation improves.

Re:Blogosphere Mood

[Anonymous Coward]

Better not blog that too bodly in your blog or the blogerati are going to get a coupla of pipe-hitting bloggers to get blog evil on your ass. Blog.

Re:Blogosphere Mood

[Geoffreyerffoeg]

What is this "special relativity"? I hate buzzwords. Just call it the ether like everyone else. I mean they're close enough...and nobody's gonna need to use "special relativity" in calculations anyway, it's just hypothetical.

Can it detect the entire spectrum of moods?

[koweja]

From whiney to really pissed off?

Re:Can it detect the entire spectrum of moods?

[mctk]

The entire spectrum of moods is from whiney to really pissed off? You're married aren't ya?

::ducks

So...

[TechnoGuyRob]

The blogosphere is female? Oh wait, we're on our way to understanding it. Nevermind then.

Mood of Slashdot?

[TheDarkener]

+1 Smart ass
+1 Pro Linux
-1 Corporatism
-1 Proprietarism (is that a word? Happy 4:20..sorry)

And...of course

+1 CowboyNeal

Disregarding the above blabber, this software sounds so flippin' cool. I've always thought about the overall mood of people, and if there's relation between it and a certain period of time. Like a website that simply asks "Are you happy right now?" or "Are you sad right now?" Very useful information! *gazes up at sky*

Tracking bad hair trends, too?

[cno3]

The software, called MoodViews, was created by Gilad Mishne and colleagues at Amsterdam University, The Netherlands. It tracks about 10 million blogs hosted by the US service LiveJournal.

Monday - Mood: Emo
Tuesday - Mood: Emo
Wednesday - Mood: Emo ...

Re:Tracking bad hair trends, too?

[zaguar]

Q: How many emo's does it take to change a lightbulb?

A: None. They just sit in the dark and cry.

Re:Tracking bad hair trends, too?

[Rachel Lucid]

I should have known Livejournal was the source of this. With people's entries naturally having a mood category, the analysis becomes next to trivial.

Here I was hoping for something clever, like word frequency or something.

Checking mood...

[JoeCommodore]

Was a yellow-green now looks like its changing to a deep purple.

Marketing Buzz Alert

[anomalous cohort]

The Blogosphere (a collective term encompassing all weblogs) isn't really addressable so how can it be measurable? It's not like there is a URL to "the blogosphere" and how would you know if you have successfully polled all blogs on the Internet? This appears to be a subtle commercial for LiveJournal.

Re:skeptical

[TechnoGuyRob]

Erm...I don't know if you read the article, but they extract the moods from LifeJournal posts, not analyze the text or anything like that.

Assuming that people are honest about their moods (and why wouldn't they be?), I don't see why this wouldn't be accurate.

Apparently, your mood right now is ignorant.

And?

[Jayjay75]

What would this tell us, exactly? That people are more inclined to get drunk on weekends and are grouchier on Mondays than on other days? This is something we don't already know?

Mixed Signals

[_Sprocket_]

Great. Humans already have trouble interpreting the tone of electronic messages. On top of that, let's have some algorithm tack on the subtle clues so necessary for proper interpretation of human communication. After all, computers have already shown a bang-up track record dealing with Human languages.

Cool project though. Hilarity will undoubtedly ensue.

Re:Mixed Signals

[patio11]

I could see this working. Is it any harder to tell if a message is, say, depressed than it is to determine if a message is a commercial pitch?* Because pure-"Bayesian" analysis of spam routinely gets 95%+ accuracy, which if we're not talking about the content of any specific message but are trying to measure trends between time periods is plenty good enough. Lets take a particular application: Apple wants to know if their iPods are still the hottest thing on the planet. Simple process: have a team of humans hand-classify 1000 posts on LiveJournal mentioning iPods (just grep for it) as "pro-iPod" or "anything else". Then have your trained classifier slurp up every post that day, discard any that doesn't mention iPod, and classify ahoy. Ten minutes later: "Beep, 87% of 23842385902 messages mentioning iPod (5.3% of all posts in last 24 hours) are mostly upbeat."

* Whether it is or not is not obvious to me, but setting up an experiment to figure out which it is isn't that hard.

Slashdot mood?

[TechnoGuyRob]

There's almost 30 posts and only one of them has been modded up (once, to funny no less)? I think I can predict the mood of Slashdot about this:

Indifferent

Or as a LiveJournalist would say:

like i don care man

Re:Reservation...

[Gyga]

I hope whatever they use listens to robots.txt

zerg

[Lord Omlette]

They did this for Livejournal. It would always, without fail, return "tired".

Re:BuuurrrPb

[Nesetril]

d00d, it's only thursday. you should feel "flirty and lonely" today not drunk and merry. you GODDAMN OUTLIER, remove yourself from the blogosphere or you will spoil our sales pitch to Google.

One word:

[Xaroth]

"Sardonic."

Re:Reservation...

[shawb]

I understand your concerns, but information that you yourself put on a webpage should not be considered private. If you don't want personal information to get out, don't post it in a public forum. Besides, it doesn't sound like this particular piece of software is really collecting personal information; all it does is look for spikes in a particular mood tag, and then parses through the text in the publically readable text to find unusual words. If this helps companies figure out what products actually excite people, then they will come out with products that people actually like (if used properly.)

As a company (and especially as an investor) I would, however, take any information gathered through such a technique with a grain of salt or two. It seems that it would be close to trivial for a company game the system and set up enough accounts or bribe enough bloggers to tip the scales one way or the other, essentially creating inter-corporate astroturphing.

What's the mood of Slashdot on this one?

[StressGuy]

"Geeky"

There I go pissing karma away again....

debating the color of Schrödinger's cat?

[Gothmolly]

Just as useless, and as unprovable. Next, specialized computer technology to detect the color underwear most bloggers wear?

Website still up?

[Chris_Jefferson]

This can't be right.. the website is still up. Perhaps that is because no-one can find the link To the actual moodviews website.

I can't decide if I should feel guilty for posting this..

Re:Website still up?

[TubeSteak]

AFAIK, they're looking at Live Journal tags and then comparing it to their analysis of the posts' text.

I'm not sure how innovative this is, without actually knowing how they guess at the mood. Maybe it's something as simple as training a Bayesian filter and then saying "gosh, look at how accurate the predictions are!" The application would be innovative, but not the method.

It would make for some great targeted advertising:
Feeling depressed: Shop online for clothes!
Feeling in love: Buy your sweet heart some flowers.
etc

i've got one of those

[syrinx]

I've found thier top-secret code:

Mood of LiveJournal: angsty

and the mood on 4/20

[SecureTheNet]

According to their page, under "Which moods are hot?" they list "high"

http://ilps.science.uva.nl/MoodViews/Moodgrapher/? high

Notice the sudden spike in the "high" mood on 4/20? I thought that was interesting.

Re:Why is it a buzzword?

[walmartshopper67]

It's a buzzword when it is used by people to describe things that have nothing to do with it. Like when 40 year old rednecks talk about one of my sites saying it is a "blog", instead of what it is and always has been, a website. A word like "slashdotting" would be a slang or scene type word, when a word like "blog" gets repeated all day in the media, usually to describe things that used to be "websites".

Been there, done that! - MoodTap.com

[bennomatic]

Why didn't this story just point directly towards the Mood Tap web site?

Horny to Depressed

[slashmojo]

Thats interesting.. it shows a spike for 'horny' followed a few hours later by a spike for 'depressed' which I presume means all those horny people didn't get laid again..

Moodgrapher

[themusicgod1]

what, like Moodgrapher? I don't understand what's so new?

Definition of blog

[DavidNWelton]

Blog \Blog\, v. t. [imp. & p. p. {Blogged} (bl[o^]gd); p. pr. &
n.

1) A foul, smelly obstruction of the sewer pipes connected to a residence. For example, "We'd better call a plumber, this blog is really bad".