2015년 3월 11일 수요일

Data Today: Machine learning, QlikView dashboard hacks + more

O'Reilly Media Logo
O'Reilly DataNewsletter

1. The secret to turning NFL players into digital gods

Here's how the Madden ratings work. (Spoiler alert: the secret is algorithmic alchemy, not magic.)

2. Machine learning done wrong

With small data, the cost of experimentation is low. But with big data it pays to analyze the data upfront and then design the modeling pipeline (pre-processing, modeling, optimization algorithm, evaluation, productionization) accordingly.

3. Year Zero: Our life timelines begin

In the next decade, Year Zero will be how big data reaches everyone and will fundamentally change our lives.

50% off algorithm books and videos (today only)

algorithm book saleLast day—save 50% on books and videos that will help you master the art of writing algorithms. Topics include learning data structures, working with algorithms in Python or JavaScript, and more.
50% off algorithm books & videos (ends today) →

4. Bridging big data silos

John Carnahan discusses holistic data analysis, engagement channels, and data science as an art form in this O'Reilly Radar podcast.

5. Better than Bloom

The authors of this paper propose a new data structure called "the cuckoo filter" that can replace Bloom filters for approximate set membership queries.
Sponsored Content

Fast data stack

voltdbWhat is fast data? It's data in motion, and it creates Big Data. But handling it requires a radically different. With the Fast Data Stack white paper from VoltDB, you'll learn how to build fast data applications with an in-memory solution that’s powerful enough for real-time stateful operations.
Download the Fast Data Stack white paper →

6. Seeing circles, sines, and signals

If you have the slightest interest in signal processing (or even a passing interest in visualizations) check out Jack Schaedler's tutorial on digital signal processing. Really, how can you resist something described as "an eccentric piece of not-so-rigorous literature with a preoccupation for explaining things using interactive visualizations, animations, and sound?"

7. Two QlikView dashboard hacks

Here's a couple tips for improving information density in your QlikView dashboard.

8. Data cleaning meets crowdsourcing

Jiannan Wang discusses human-in-the-loop machine learning—particularly in the area of data cleaning. Interested in machine learning? You might also want to check out the free report Real-World Active Learning by Ted Cuzzillo.

Strata + Hadoop World Startup Showcase deadline

Strata + Hadoop WorldWant to get your startup in front of some of data's brightest minds, as well as potential customers and funders? The Startup Showcase at Strata + Hadoop World in London is a great opportunity to show off what you've got. But the application deadline is 3/25, so you'll have to hurry.

Reminder: The Early Price for Strata + Hadoop World London registration ends next week. Reserve your spot now.

9. The 3 degrees of Eric Roberts

The ubiquitous villain, Eric Roberts, has supplanted Kevin Bacon as "the center of the Hollywood universe." 25% of the 1.91 million actors on IMDb are within 2 degrees of separation from Eric Roberts. By 3 degrees out, Eric Roberts can be connected to 88% of all actors. (If he had ever appears in a movie with Kevin Bacon, his dominance will be complete). Here's a data vis to show why the 6 degrees of Kevin Bacon game should now be the 3 degrees of Eric Roberts game.

10. Freebie of the week

free report Real-world HadoopUsing Apache Hadoop, Apache HBase, and related NoSQL technology is a cost-effective way to quickly get value from data. This free ebook, Real-World Hadoop,shows you how to use Hadoop and NoSQL successfully. Co-authors Ted Dunning and Ellen Friedman share lessons taken from real-world situations, including examples of Hadoop and NoSQL in production settings. Thanks to MapR for this week's freebie.
Get the free ebook →

댓글 없음:

댓글 쓰기