News for hackers. Stuff that matters.
Re:The Most Innovative Companies In Big Data by genofon on Saturday March 29, 2014 @10:03AM (#7492243)
the right title should be: "The World's Top 10 Most Innovative Companies In Big Data according to their marketing team." The article I would like to read is about the real advantages they made, not just "they are applying BIG DATA to the X,Y industries". No description on the size of the problem nor the advantages or details.(I'm especially looking at you AYASDI... )

I've been to Big Data conferences where the main application was mean and standard deviation on a huge datasets, I'm just sick of companies inflating the Big Data bubble (and this comes from a "data scientist")

Re:The Most Innovative Companies In Big Data by michaelochurch on Saturday March 29, 2014 @10:35AM (#7492344)
I've been to Big Data conferences where the main application was mean and standard deviation on a huge datasets, I'm just sick of companies inflating the Big Data bubble (and this comes from a "data scientist")

"Data science" is a bizarre job description to me. Some companies' data science teams are doing work that could be done in Excel, and others are doing sophisticated machine learning research. In many companies, though, it's a watered-down version of the (extinct, sadly) R&D job description. I've heard people opine that 95% of "data scientists" have never implemented anything more sophisticated than an SGD regression, and it wouldn't surprise me.

Granted, you can get some neat insights and visualizations with off-the-shelf tools, but I still consider it important to know how the algorithms actually work and what the basic assumptions are. For example, you can usually use linear regression for 2-class classification problems (as opposed to the mathematically more correct logistic model, since probabilities are in [0, 1] and linear models diverge at the boundary) get a reasonable predictive model, but it's worth knowing when (and why) that short-cut breaks down.

In software, you hear "data scientist syndrome" used to describe people who have a lot of short job tenures because they (a) leave if there isn't interesting work for them, and (b) tend to be the first laid off, not because they're bad but because R&D is first to bleed when things go bad. The software industry forces you to choose between short-term job security (people doing interesting work are most exposed to organizational changes) and long-term career health (people not doing interesting work turn into dinosaurs). It stands to reason that the selection process for organizational credibility (favoring tenure) would be something other than deep knowledge of data science (which requires a stream of interesting work, and the behavioral correlates such as job volatility).

It's sad that the typical corporate culture of reliable mediocrity, at the expense of excellence, has also learned to use the vocabulary of "Big Data". Reality: truly Big Data (> 10 TB) is a huge pain in the ass. It's what you have to deal with when there is too much noise, or the model's essential complexity is too high, to get a good model out of "small data".

Re:The Most Innovative Companies In Big Data by genofon on Saturday March 29, 2014 @01:32PM (#7493024)
-Data science" is a bizarre job description to me. I totally agree, that's why I've quoted it, it's one term that everyone use without knowing the meaning
Re:The Most Innovative Companies In Big Data by dannypgh on Saturday March 29, 2014 @10:47AM (#7492382)
Curious- why would you ever use linear regression for binary classification?

Logistic regression can be implemented as linear regression predicting the input to the logistic function.

Re:The Most Innovative Companies In Big Data by oldskoolbob2000 on Saturday March 29, 2014 @10:55AM (#7492413)
From my experience, it's easier to explain linear coefficients. Also, at least in R, linear regression tends to run faster than logistic regression.
Re:The Most Innovative Companies In Big Data by sgy on Saturday March 29, 2014 @12:08PM (#7492688)
Re:The Most Innovative Companies In Big Data by michaelochurch on Saturday March 29, 2014 @01:31PM (#7493023)
Linear regression has a closed-form solution and, even when that is impractical (50k+ features, at which point you start not wanting to do the O(p^3) operations involved) and you have to use gradient descent, it still tends to run faster.

If most of your data are in the p ∈ (0.2, 0.8) range, then the logistic curve is approximately linear anyway. Where linear regression fails you on a logistic problem is if you have a lot of data where the linear model will give p's outside of (0, 1).

While the linear model's mathematical foundations are "incorrect"-- more precisely, it's not the maximum-likelihood model and may have a conditional likelihood of zero, that is, being impossible as "the correct" model on the data-- the truth is that, at 100k+ features, getting "the right" model is effectively impossible with most real-world data sets and you have to use simplifying assumptions and techniques (e.g. regularization, early stopping) anyway. Linear regression doesn't usually do as well as logistic regression, but sometimes it can, and most people are going to try both.

Re:The Most Innovative Companies In Big Data by dannypgh on Saturday March 29, 2014 @05:02PM (#7493918)
I think I see the source of the confusion. Linear regression does not imply any specific method for solving for your coefficients. You can solve a linear regression by inverting your matrix or you can use gradient descent, to give two mechanisms.

Yes, solving for your coefficents based on inverting the matrix can be O(n^3) (although wikipedia tells me there's ways of doing it that are O(n^2.373), but at the end of the day, it's slow for large n) but gradient descent doesn't have this problem, as gradient descent is O(n) with regards to your training set.

And if you're implementing gradient descent to build a classifier, you can equivalently do logistic regression by just changing your cost function (that is, you'll be solving a linear regression to give you f(x) such that S(f(x)) is the probability you care about, with S(x) being the sigmoid function).

So linear regression can perhaps be easier than logistic regression, but only when n is so small that you're solving your linear regression by inverting the matrix. Once you've switched to gradient descent (or anything that can solve for solutions to arbitrary cost functions) there is no difference.

Re:The Most Innovative Companies In Big Data by michaelochurch on Saturday March 29, 2014 @05:45PM (#7494127)
Once you've switched to gradient descent (or anything that can solve for solutions to arbitrary cost functions) there is no difference.

I'd have to work this out on paper. I think that's essentially right, but most of your CPU cost is going to be in evaluating that function (or, more accurately, its derivative) and not all functions are, in terms of computational cost, equal. I think the logistic model might take longer to train, but I'd have to work it out to be sure.

You're right that the code isn't much different.

So linear regression can perhaps be easier than logistic regression, but only when n is so small that you're solving your linear regression by inverting the matrix.

The matrix grows with the number of features, so if you have 1 billion observations but only 100 features (dimensions) you can do that.

Re:The Most Innovative Companies In Big Data by dannypgh on Sunday March 30, 2014 @12:33PM (#7496684)
I doubt the CPU costs will be measurably different.

I suggest watching: https://class.coursera.org/ml-003/lecture/36 and you can jump to 7 minutes if you don't want to watch the whole thing.

The only difference is instead of merely doing the matrix multiplication of your observation times your coefficients (the scalar product of θ and x) you have to take the output and use it as the input to the logistic function. The matrix multiplication will grow in complexity with the number of non-zero features (either non-zero coefficients in your model or non-zero features in your observation, presumably the latter) and the evaluation of the sigmoid function will be constant time, and on many architectures will be a very small number of instructions.

If you're operating in very small numbers of dimensions what you're doing may still be interesting but I'm not sure it qualifies as "Big Data," regardless of the number of available observations. After all, if you're producing something with actual utility, you'll probably converge sooner rather than later...

Re:The Most Innovative Companies In Big Data by ForHackernews on Saturday March 29, 2014 @05:45PM (#7494128)
"Data Science" is when you do statistics on a macbook.

https://twitter.com/BigDataBorat

Re:The Most Innovative Companies In Big Data by bsg75 on Saturday March 29, 2014 @12:56PM (#7492883)
>"Data science" is a bizarre job description to me.

It is starting to smell like a marketing term, where "Big Data" is used to get executives with purchasing authority to be interested in no-longer-new tech by putting a shiny wrapper on it, "Data Science" is doing something similar with job titles - perhaps in an effort to get those who know better to buy into the hype.

Re:The Most Innovative Companies In Big Data by kyllo on Sunday March 30, 2014 @01:20PM (#7496937)
"Data Scientist" = statistician who knows how to use computers.
Re:The Most Innovative Companies In Big Data by sgy on Saturday March 29, 2014 @12:04PM (#7492674)
Mean and standard deviation are [measures] to calculate data, but not [applications]. Application is making that data meaningful and organized, in order to push whatever industry forward. It's just what actually IBM, GE et. all. are doing.

Still, the article doesn't have to be objective and you might be right but a little bit negative.

Re:The Most Innovative Companies In Big Data by ghoul2 on Saturday March 29, 2014 @10:54AM (#7492406)
eh. not including Google in a list of big data companies is of similar kinda ridiculousness as not giving the Nobel Peace prize to Gandhi. It doesn't take anything away from Google/Gandhi, it just marks the list/prize as farcical.
Re:The Most Innovative Companies In Big Data by Fishkins on Saturday March 29, 2014 @12:08PM (#7492691)
Just for context, Google was #1 on their general "most innovative" list. I think it's pretty common to try to have a variety of companies topping different lists, even if one company really should be at the top of them all. That doesn't make this list any more accurate, but it might have been what they were thinking.
Re:The Most Innovative Companies In Big Data by coldcode on Saturday March 29, 2014 @11:00AM (#7492435)
What does "Big Data" even mean? Personally I think of it as some kind of disruptive synergistic paradigm. Is it a sea change or a holistic approach? Perhaps some kind of seamless mission-critical win-win? I hate when people use overloaded terms that can be twisted to mean anything at all. To not put Google in this list clearly shows it's just a random collection of companies, or maybe based on advertising sales to the magazine.
Re:The Most Innovative Companies In Big Data by oldskoolbob2000 on Saturday March 29, 2014 @11:10AM (#7492460)
Well said. To me, the definition of 'Big Data' is data so big that that the computation time is much greater than the data analysis/munging time. It has nothing to do with how complex, robust, etc. the analysis is. Big data is really not that sexy, but it's a fad that I'm riding because people are hiring for data analysis skills.
Re:The Most Innovative Companies In Big Data by jgalt212 on Saturday March 29, 2014 @08:38AM (#7492045)
I am bit surprised to see both Palantir and Factual missing b/c 1. they are doing interesting work, and 2. they probably spend a pretty penny on PR.
Re:The Most Innovative Companies In Big Data by walshemj on Saturday March 29, 2014 @03:03PM (#7493474)
Lexis Nexis Risk (HPCC) and Cloudera
Re:The Most Innovative Companies In Big Data by michaelochurch on Saturday March 29, 2014 @10:24AM (#7492304)
Factual has smart people at every level (a rarity in the "tech" world) and seems to be able to market itself organically to top developers. They have an extremely competent core team that people want to work with. This doesn't require a PR budget.

It's the Knewtons of the world that have to spend millions on PR. Actually, Knewton is solidly OK at hiring good people but they usually leave within a year because of the management.

Re:The Most Innovative Companies In Big Data by S4M on Saturday March 29, 2014 @12:07PM (#7492686)
I noticed several of your comments criticizing Knewton. Could you expand a bit? I am not related to Knewton at all, but interested in ed tech. If you don't want to say it in public, I'd appreciate if you could email me.
Re:The Most Innovative Companies In Big Data by michaelochurch on Saturday March 29, 2014 @01:07PM (#7492934)
I'll post half of it publicly and email you the other half.

Their management is known throughout the industry to be unethical. Some recruiters refuse to work with them, and you'll hear some amazing stories if you hang around New York. They've existed for almost 6 years, pivoted constantly and recklessly, and delivered next to nothing. They're great at using the founder's family connections to raise money and get partnerships, but they treat engineers like commodities and have extreme architectural instability. There's also one high-profile case of the execs spending months trying to ruin the reputation of someone after he left.

Ultimately, their true mission is fire teachers. That's a really terrible mission, if you ask me. Obviously, the engineers and data scientists (Knewton's management is scum, but they have good people at lower levels) are being told a different story about "democratizing education", but their real mission is to make teachers obsolete while making a few ed-pub incumbents (like Pearson) very wealthy.

They're really damaging the reputation of the ed-tech space. It's a shame, because there are some really good companies trying to advance the field, and they're struggling to raise money due to the behavior (and declining reputation) of a large no-longer-startup that has nothing to do with them.

Re:The Most Innovative Companies In Big Data by sgy on Saturday March 29, 2014 @09:06AM (#7492114)
"interesting work" is pretty much abstract.
Re:The Most Innovative Companies In Big Data by jgalt212 on Saturday March 29, 2014 @09:18AM (#7492140)
keep on with those trollish sub-comments. It will do wonders for you 0.7 average karma per post.
Re:The Most Innovative Companies In Big Data by sgy on Saturday March 29, 2014 @09:28AM (#7492161)
thank you for the advice. http://ycombinator.com/newswelcome.html
Re:The Most Innovative Companies In Big Data by jgalt212 on Monday March 31, 2014 @10:40AM (#7501707)
will you and your loser friends get out of your mother's basement for a few minutes and stop downvoting me?
Re:The Most Innovative Companies In Big Data by dang on Wednesday April 02, 2014 @03:15AM (#7514056)
Totally inappropriate for Hacker News and guaranteed to get anyone banned who does it repeatedly.
Re:The Most Innovative Companies In Big Data by capkutay on Saturday March 29, 2014 @01:32PM (#7493025)
Excuse my marketing terminology but I think splunk was only successful because they were the only enterprise "big data" player present on the way to the peak of inflated expectations[0]. Technology wise, it's easy to poke holes in and not even valuable with free indexing tools like elasticsearch[1] coming out and being paired with logstash. And if you think a distributed inverted index for syslogs is the future of real time analytics, you're probably stuck in 2009.

0: http://berglondon.com/wp-content/uploads/2010/07/trough.png

1: http://www.elasticsearch.org

Re:The Most Innovative Companies In Big Data by beamer99 on Saturday March 29, 2014 @09:16AM (#7492135)
The slant in the list is companies doing interesting things with big data or applications from unexpected areas like weather and mapping it onto things like shopping habits. What I don't understand is how Splunk is on the list since operational data and business intelligence has been a historical fit for big data. Nothing new. IBM is evangelizing big data. A bit of a weak one, I thought but the Smarter Cities initiatives doing some good work. Some interesting companies on the list and in the comments section.
Re:The Most Innovative Companies In Big Data by theyeti on Saturday March 29, 2014 @08:32AM (#7492028)
A bit surprised not to see Google in that list.