« Social Media Analysis Stampede | Main | Push Button Politics »
July 03, 2007
Comments
The comments to this entry are closed.
March 2016
| Sun | Mon | Tue | Wed | Thu | Fri | Sat |
|---|---|---|---|---|---|---|
| 1 | 2 | 3 | 4 | 5 | ||
| 6 | 7 | 8 | 9 | 10 | 11 | 12 |
| 13 | 14 | 15 | 16 | 17 | 18 | 19 |
| 20 | 21 | 22 | 23 | 24 | 25 | 26 |
| 27 | 28 | 29 | 30 | 31 |
Recent Posts
- AI's not AI
- Agile Web Mining at Bing
- The Economist gets in on the AI Fluff
- AI, Artificial Birds and Aeroplanes
- How to Understand Computers in Film
- How the Tech Media Keeps Artificial Intelligence at a Distance
- How Hollywood Keeps Artificial Intelligence at a Distance
- Artificial Intelligence and Economics
- Stamp of Approval
- The Longform Manifesto
Categories
- aaai2007 (2)
- advertising (7)
- affect (1)
- AI (22)
- alexa (2)
- aol (1)
- ask (3)
- astronomy (1)
- attenex (2)
- attensa (1)
- attentio (1)
- bbc (13)
- bing (2)
- blogburst (3)
- bloglines (5)
- blogosphere (218)
- blogpulse (17)
- blogs (89)
- boards (26)
- boardtracker (9)
- boingboing (1)
- Books (12)
- browserWars (1)
- buzzlogic (2)
- buzzmetrics (1)
- calendaring (1)
- china (1)
- clustering (4)
- CoComment (1)
- colbert (2)
- comments (1)
- computer graphics (1)
- Conferences (3)
- Content (5)
- d8taplex (33)
- data engines (43)
- datamining (75)
- DataMiningElements (2)
- datasphere (2)
- dataviz (208)
- dataweb (8)
- dell (1)
- design (3)
- digg (1)
- document analysis (8)
- economics (7)
- edgeio (7)
- email (3)
- facebook (2)
- farecast (1)
- fastreams (3)
- featuredhyperlocalpost (1)
- feedburner (4)
- Film (44)
- findory (1)
- flickr (7)
- followthemoney (1)
- foo (3)
- Games (5)
- geocoding (24)
- gigapan (2)
- GIS (130)
- GIS data (12)
- google (36)
- graphs (46)
- GUI (11)
- history (2)
- hoosgot (1)
- html5 (3)
- humour (1)
- hyperlinkedsociety (3)
- hyperlocal (24)
- icerocket (1)
- icwsm (37)
- icwsm2008 (3)
- icwsm2011 (1)
- image exploration (2)
- infographics (4)
- informationextraction (4)
- intelliseek (4)
- iPad (1)
- iphone (10)
- katrina (2)
- kdd (1)
- klostu (3)
- lijit (1)
- linguistics (7)
- linkrank (5)
- listing (1)
- liveearth (1)
- livejournal (2)
- livelabs (13)
- local (16)
- longtail (7)
- machinelearning (4)
- marketingIntelligence (42)
- mashups (2)
- measuremap (2)
- memeorandum (3)
- memetrackers (8)
- microblogging (1)
- microformats (3)
- microsoft (10)
- mobile (5)
- Music (4)
- nasa (3)
- news (61)
- nlp (51)
- notfree (2)
- nsa (1)
- online personal media (4)
- opendata (4)
- OS (1)
- oscars (2)
- pdf2008 (3)
- philosophy (5)
- photosynth (10)
- podcast (1)
- podcasting (5)
- politics (49)
- powerset (11)
- predictions2006 (4)
- pronto (1)
- psychology (2)
- pubsub (1)
- ranking (4)
- recovery2 (1)
- reddit (1)
- reverbiage (1)
- rojo (1)
- rss (36)
- Science (4)
- search (234)
- search-problems (14)
- semanticweb (1)
- sentiment mining (46)
- site statistics (6)
- sketchup (2)
- social software (3)
- socialmedia (150)
- socialnetworks (19)
- software (1)
- sotb (1)
- space (1)
- spam (14)
- sphere (7)
- spinn3r (1)
- spongecell (3)
- Sports (4)
- spotplex (1)
- statistics (7)
- superbowl (4)
- swivel (2)
- tableunderstanding (1)
- tag (2)
- tagging (7)
- tagnorati (1)
- tags (6)
- tagtastic (1)
- tailrank (10)
- technorati (36)
- Television (3)
- text mining (47)
- timeseries (7)
- topix (1)
- Travel (3)
- turn (1)
- twitter (45)
- typepad (4)
- umbria (2)
- video (1)
- vivisimo (3)
- VoIP (1)
- web2.0 (10)
- WebApplications (3)
- Weblogs (6)
- wikipedia (3)
- wired (2)
- worldcup (5)
- yahoo (1)
- youtube (11)
- zerotolerancesearch (1)
- zonetag (1)
Although the announcement peak was larger, we'd expect that announcements have more of a peak than the product release, which has people looking for reviews and specs over a longer period of time. The human eye though picks up on peaks. What would be a better way to display the data to emphasize overall volume instead of peaks? My first guess would be to use larger time windows, but moving averages might also work.
Posted by: Amit Patel | July 04, 2007 at 09:21 PM