I'm a co-founder at EffectCheck and I was working closely with Scott this weekend as he was building this. It wasn't really ready for viewing yet, but okay... :)
Please note that the top graph is currently a mixture of two sets of data. The older points were using a less sensitive and improperly calibrated HN comment model, hence why everything is drifting around near "Typical." The points from 14:47 onward are using the correct model.
For those wondering why HN was so negative from 14:47-20:47, I believe the main topics of conversation were the Bitcoin and Sega debacles. Makes sense that people were really anxious given all that news.
Also, allowing them to see the popularity of certain trends (traffic, if you will). I am not familiar with the industry…maybe the product exists…but it seems incredibly useful!
Well done, too.
http://rawkes.com/blog/2011/04/19/finding-patterns-in-twitte...
1. Depresssion 2. Anxiety 3. Hostility 4. Happiness 5. Confidence 6. Compassion
Anxiety and Depression definitely go together, in the same way that hostility and happiness are actually similar in terms type of emotion, in terms of level of intellect, and confidence and compassion would be at the highest end of that spectrum.
I would also change the colors to group them: anxiety and depression as blues, hostile and happy as reds, and confidence and compassion as greens. Or something like that.
My co-founders [1][2] actually did spend a decent amount of time thinking about this. The emotions are first sorted into negative on the left and positive on the right. The order is for symmetry of left/right side words:
- Anxiety is the opposite of Confidence
- Hostility is the opposite of Compassion
- Depression is the opposite of Happiness
As for the colors, each one is correlated to the typical psychological association for that color. The exception is Happiness which should be yellow but yellow doesn't render well on websites.
[1] Yanon Volcani (Clinical Psychologist): http://www.volcani.com
[2] David Fogel (AI Expert): http://www.natural-selection.com/people_dfogel.html
In any case, anxiety and depression are generally a package deal in clinic. So I would put them together visually.
Depression --------------------X------- Happiness
Anxiety ----------------X----------- Confidence
Hostility -------------------x-------- Compassion
It can be used for a stock analysis and dampening or amplification caused from other firehose like sources like Twitter.
We are focused on B2B applications of our technology rather than the consumer/API side. However, if you have a cool idea for how you'd like to use EffectCheck, email me [1] and I'll be happy to discuss it with you.
[1] wesley.tansey@effectcheck.com
It was neat when it came out years ago... Although still not sure how to make any meaningful use of this tool. I suppose it is more useful for content generators..
Anxiety: #2288ff vs. #595884 (anxious)
Hostility: #ee0000 vs. #E0192C (angry) -- hostile only had 100 samples
Depression: #3300cc vs. #283152 (depressed)
Confidence: #ee00cc vs. #FF7F00 (confident)
Compassion: #11bb00 vs. #00696F (sorry) -- not sure on this word choice but compassionate wasn't available. Loving was #004E6F
Happiness: #ff8800 vs. #FF7F00 (happy)
So I would say we did well on all except maybe Confidence. However, confident shows up as almost the same color as happy, so we would have to differentiate somehow anyway.
[1] http://blog.effectcheck.com/2011/05/31/do-social-news-sites-...
