via Planet Python:

Given a table like this:

mysql> select * from delta;

+------------+-------+

| end_time   | level |

+------------+-------+

| 1245937320 |     5 | 

| 1245937565 |     1 | 

| 1245954231 |     5 | 

| 1245954844 |     2 | 

| 1245956014 |     4 | 

| 1245963214 |     5 |

... [-->]

I've seen people drawing ascii-charts with sql-statements, but this is new.

via Data Mining: Text Mining, Visualization and Social Media - (Matthew Hurst):

Marti Hearst - pre-eminant in the fields of text mining and user interfaces - yesterday published her book on Search User Interfaces. I intend to write up a review of the content later, but for now heres a summary.

The book, which is available ... [-->]

Just skimmed the online version of the book.

via Recent Google Publications (Atom) - (D. Sculley, Gordon V. Cormack):

Going Mini: Extreme Lightweight Spam Filters, D. Sculley, Gordon V. Cormack [-->]

Nice result. I wonder if it's really necessary to train only on the top features. Why not train on all features and then select the top weigths for a smaller model? If the goal is only to be able to store all models in memory, that should be enough.

via neingeist. borschtsch.:

[Reposted from arduino via [-->]

Haha, sehr schön - Film@11. Klingt gar nicht so absurd, wie die ganzen "Rettet unser obsoletes Geschäftsmodell!"-Aufrufe von heute.

via wirres.net - (ix@schwenzel.de (ix@wirres.net))

bundestag:~ uschi$ inaptitude install –skip-gewaltenteilung 

–ignore=petition –ignore=experts –force zensurinfrastruktur

[geklaut bei feint.subsignal.org] ... [-->]

@bundestagswahl$> apt-get dist-upgrade

via Google Research Blog - (Research Admin):

Posted by Jake Brutlag, Web Search Infrastructure

At Google, we've gathered hard data to reinforce our intuition that "speed matters" on the Internet. Google runs experiments on the s ... [-->]

The surprising fact for me is, that slowness has so little impact on usage. I would have expected that it's much more important. Maybe this is because google owns a quasi-monopole on websearch... waiting 400ms for a search request seems pretty much unacceptable for me.

.oO(someone named brutlag doing research on internet latency... bad joke or bad luck?)

via nameless:

Grünabfälle?

[Reposted from [-->]

s/^s/g/