Comments
Finding Answers Through Prescriptive Analytics
You must login to participate in this chat. Please login.

So are the areas we mentioned (descriptive, predictive, prescriptive, automation) simply levels of human vs. system in decision-making?

Statistician

MLFoGrasso - I think Moneyball started to provide a set of predictive projections on what could happen to players. it was far from perfect, but at least gave the A's enough of a starting point to attempt to do arbitrage on future-facing outcomes. Baseball has enough of a scouting component that they're still a while away from truly prescriptive approaches.

Prospector

Thanks everyone. Have a great day.

Editor

Prescriptive analytics is based upon assumptions that lie in predictive analytics and descriptive analytics, and predictive analytics itself is based upon assumptions that lie in descriptive analytics.  So MB was about changing the assumptions.

Blogger

I'd love to connect with you on Twitter at @hyounpark or feel free to email me at hpark @bluehillresearch.com if you have any additional questions. Thank you for the conversation!

Prospector

@Joe I see that.  I guess it's just that someone had to make the decision at some point about which players to sign/release, and how much did the MB approach guide that decision?

Statistician

There were still humans making the decisions.

Blogger

@MFL: Moneyball was really about redefining descriptive analytics.

Blogger

Hear, Hear It was very informative !    Thank you both.....

Blogger

Thanks, everyone for a good discussion. Great job, Hyoun. Would you like to share your social media info again?

Editor

Since baseball has come up, moneyball = prescriptive analytics?

Statistician

Just wanted to say thanks to both Hyoun and Jim for engaging discussion...

Blogger

That pharma app sounds interesting. Hyoun, we are approaching the top of the hour. So, I welcome any final comments or questions.

 

Editor

As for fantasy baseball, a truly prescriptive analytic solution would make sure that I never drafted any Yankees...

Prospector

This is a personal bias, but I love prescriptive analytics that are used to identify optimal drug design for personalized drugs, meaning that the drug has been designed both to solve a medical problem and is optimized for personal chemistry and other drug interactions. It's amazing that what used to be an endlessly iterative process with dead end after dead end can now develop an actual best recommendation based on thousands of interactions.

Prospector

Gee whiz as in good, surprisingly good

Editor

Is prescriptive analytics being applied in...

* public transportation planning decisions?

* urban planning decisions?

Blogger

@Jamescon

"Geewhiz" as in good or as in bad?

Prospector

"Question I have for our speaker: Given that prescriptive analytics can only be as good as the logical sensibilities, access to and knowledge of factual context, and similar factors, what flexibility exists in present and near-future prescriptive analytics tools to allow data scientists in an organization to modify the prescriptive analytics tools?

And, more to the point, how do you keep them from screwing it up with their "human" biases/failures?  ;)"

 

Great question. The level of contextualization associated with decision-making is the real differentiator. We're trying to get context that matches human insight and reasoning. Of course, the flipside is that when machines can make human decisions consistently, we have a separate problem...

Prospector

So it's not particularly impressive.  But it, in a way, qualifies.

Blogger

@James: It does.  I used to be big into fantasy baseball several years ago.

But no, it's just a really basic system.  You rank your players in order.  The automated program just picks from your list top to bottom of players still available.  I think there may be some basic, not very intrusive override feature just to make sure you get all your positions, but that's it.

Blogger

Hyoun. You cited several realworld uses for prescriptive today. Would any fall into the "gee whiz" category for you?

Editor

Joe. Does Yahoo do that now on baseball? Does it work any better than the analytics used  by a certain Boston baseball team?

Editor

@Joe  Interesting example , who knew Yahoo was so far ahead of the curve when it come to prescriptive analytics !

Blogger

"It seems to me that the most rudimentary prescriptive analytics-cum-automated system that comes to my mind immediately is an automated Yahoo! fantasy sports league draft"

Assuming that you trust the Yahoo rules, that's spot on. The challenge occurs when Yahoo auto-chooses someone you would have never picked for yourself. That's where the pre-analysis data assumptions end up being so important.

Prospector

You have the list of players from top to bottom, and the system makes decisions for your draft based upon what happens in the draft and who is still available in each of your turns.

Blogger

Prescriptive analytics provide a specific answer to a specific problem. In a world where we had robots with human-level motion, articulation, and mobility, we could automate the recommendation from prescriptive analytics.

Prospector

It seems to me that the most rudimentary prescriptive analytics-cum-automated system that comes to my mind immediately is an automated Yahoo! fantasy sports league draft.

Blogger

That "rules" aspect is interesting, considering that strictly human based decision making so often is based on pure rules, including the popular "we've always done it that way"

Editor

"Imagine if we could effectively combine predictive analytics, prescriptive analytics, and automation with social media?  Twitter would be Guy Kawasaki's robot talking with Robert Scoble's robot all day.  ;)"

That would be scary. The Guy robot would know exactly what to say to push the Scoble robot's buttons, as well...

Prospector

We've been kicking around the concept of the number of options that a system may present and the likely success factors. That sounds to be critical

Editor

Thanks for the great discussion so far. One of the key aspects of prescriptive analytics I didn't mention is that we're limited so far in what we can do there simply because of our trust of automated systems. There's only so much that we currently trust "rules" to do without human context.

Prospector

But sounded to me like "automation" was the next step -- routing the "prescription" straight into the robot...

Blogger

@Jamescon: One game where I *don't* like leveling up.

Blogger

Welcome Hyoun. Great presentation. thanks

Editor

Lyndon. We already have levels of death. It's called aging

 

Editor

@Lyndon I don't think anyone is suggesting removing the human element from analytics, so context should certainly play a role in any decision.

Statistician

Sorry for the delay!

Prospector

In the future, even levels of death may be an option...

Blogger

Jim says 

==

You folks are right, a single, hard answer could be tricky.

==

Sounds like some prescriptive analytics recommendations ("prescriptions"?) need to be "contextualized"?

Blogger

Hyoun should be joining in a minute

Editor

@jim   Agreed I think most decisions are based on degrees unless you are talking about death or taxes. : ) 

Blogger

I could be mistaken, but I remember a demo suggesting that it did.  So I imagine other tools on the market do as well.

Blogger

@Jamescon: I believe Watson already does this.

Blogger

@Louis: It would probably improve Facebook.  :p

Blogger

@Joe   Too much for the World of Twitter I would think and it just might bring FaceBook down completely !  : ) 

Blogger

I think if someone is going to raise the topic of prescriptive in their organization to the business side, one of the key things to emphasize will be that the system offers several suggestions with probabilities of success. You folks are right, a single, hard answer could be tricky.

 

Editor

Thanks Hyoun for explaining a very difficult concept for many......

Blogger

Welcome everyone. Apologies for the occasional static.

Editor

Imagine if we could effectively combine predictive analytics, prescriptive analytics, and automation with social media?  Twitter would be Guy Kawasaki's robot talking with Robert Scoble's robot all day.  ;)

Blogger

And now the chat-only portion begins!

Blogger

@Joe- Turn that into a meme with a picture of a fan. :)

Editor

@MFLoGrasso   Indeed.    But I for one will remember this approach best !  : ) 

Blogger

LOL @Joe.  Why do I think "stuff" is not the word many of us have going through our heads.

Statistician

Here's my understanding:

Descriptive analytics: Stuff happens.

Predictive analytics: Stuff is about to hit the fan.

Prescriptive analytics: Do this stuff.

Automation: Let me take care of all that stuff for you.

Blogger

That is good to know - prescriptive analytics needs to be quantatively based.    Don't place too many variables and unknowns into it.

Blogger

So realistically, what's the difference from prescriptive analytics and automation? I mean, obviously, some analytics don't just turn a switch or something, but do you just automatically follow the prescriptive analytics like an autopilot?

Editor

often the result of any analysis does not render just one, single recommendation, but multiple options -- maybe 2-3 top choices with probably success estimate for each

Blogger

Have they already discussed the issue of rendering more than one "prescription" from the analytics modeling?

Blogger

So obviously every type of analytics is important at different times. What is a good best practice for employing the right kind of analytics for the right project?

Editor

So we only have a raw statistic with little context.

Blogger

First refresh didn't do a thing, but I just refreshed again and all seems well...

Blogger

It is only the host's audio that is breaking up. Hyoun is fine.

Blogger

Does Prescriptive Aanalytis have different analytics methods than Predictive Analytics? 

Prospector

@daba1234    lol....try a refreash

Blogger

My audio player just keeps recycling "Lock Talk Radio" over and over....

Blogger

The problem with the basketball analogy is that statistic only measures how often the player makes the shot when the player decides to take a shot.  It doesn't demonstrate when the player will be in a good shot-making position, and it doesn't tell us how much of that 50 percent is because of the player's good shot-making positioning and how much of that 50 percent is due to the player's good judgment in assessing when his positioning is good and how much of that 50 percent is due to his actual shot-making skill and type of shot-making skill.

Blogger

Apologies for those with audio problems. I'm having no issues with the guest audio so hopefully it means the archive will be clean if you need to try later.

Editor

all the big data in the world didn't save this audio cast from 1999

Prospector

@Lyndon  Yes  me too - finally stopped on refresh.......

Blogger

Mine just rebooted too. Can't hear anything now.

Prospector

this is nothing but trouble from the audio for me too. I think the speaker is in a car going through tunnels.

Prospector

@Joe   I see.   Watson is a good example for understanding.

Blogger

@Lyndon Why not? Doesn't a medical doctor often have multiple options to help you relieve your health concerns?  But his expertise leads him to choose what prescription to attempt first.  Similarly, the analyst's expertise gives him the the wisdom (hopefully) to choose the best of the available options.

Statistician

Yes, my audio is rebooting

Prospector

My audio player just keeps rebooting. Anyone else having audio problems?

Blogger

@Louis: More or less.  Just like with predictive.

Blogger

Joe, yes. A decision support system can help human make that decision. However, the data which is the base for the decision is that generated by predictive analytics.

Data Doctor

Could prescriptive analytics render more than one choice of a course of action? i.e. more than one "prescriptions"?

Blogger

@Louis: Watson does offer a certain degree of prescriptive analytics.

Blogger

So is it just boiling down probabilities ?

Blogger

Most prescriptive analytics offer multiple recommendations with potential outcomes as the ultimate results or optimized outcomes. Usually it is not one recommendation, right?

Blogger

@Waqas/David: Engineers do tend to have an ego.  ;)

Blogger

@waqasaltaf- Fair enough. But there's a real issue when the data and the management experience don't match.

Editor

Prescriptive analytics, on the other hand, do much of that decision-making work for me (or, at least, helps with that decision-making through recommendations).

Blogger

David, managers who are open to criticism don't over-react to dissenting opinions. Others make it an ego issue.

Data Doctor

In other words, predictive analytics may tell me what's going to happen in the future, but I still have to make a human decision -- which may or may not be flawed -- regarding what to do about that future event that my analytics are predicting.

Blogger

What tools are going to tell you the best course of action ?

Blogger

I was wondering why I have not heard of this term, apparently IBM coined it (literally) in 2003 ?

Blogger

@Lyndon: Predictive = X will happen.  Prescriptive = Do Y.

Blogger

It seems like managers have a problem when you tell them there is one best answer especially when it differs from their answer.

Editor

That may be how it's supposed to work, but it's an optimistic view of healthcare.  ;)

Blogger

Need to clarify difference between PREDICTIVE and PRESCRIPTIVE analytics

Blogger

The host is breaking up. Hyoung sounds fine.

Blogger

Question I have for our speaker: Given that prescriptive analytics can only be as good as the logical sensibilities, access to and knowledge of factual context, and similar factors, what flexibility exists in present and near-future prescriptive analytics tools to allow data scientists in an organization to modify the prescriptive analytics tools?

And, more to the point, how do you keep them from screwing it up with their "human" biases/failures?  ;) 

Blogger

OTC analytics, the next frontier!

Blogger

I wonder how long after analytics are prescriptive can we get them over the counter. :)

Editor

great topic today

Prospector

Hi, folks.  Looking forward to this.  :)

Blogger

Hi all! 5 minutes to show time.

Editor

Hello Everyone !  Looking forward to the show !

Blogger

..

For what it's worth, I posted a bit of background about Blue Hill Research and prescriptive analytics in a comment: http://www.allanalytics.com/messages.asp?piddl_msgthreadid=282535

Blogger

Looking forward to the show.

Data Doctor

Folks. If you have questions for Hyoun you can post them here in the "Your Post" box. It's also a great place to share comments with your peers in the audience.

Editor

45 minutes to showtime, ezAllAnalytics

Editor

I see. It is not yet show time. Sorry.

Prospector

Has it started? Should it show up on this page?

Prospector

Since you are logged in, you will hear the audio when it starts at 2 pm. If it doesn't start for you at that point try refreshing your browser.

Editor

how to hear the Audio?

 

Prospector

Welcome, Kalimko. We are about an hour and 15 minutes away from the start of the show.

Editor


INFORMATION RESOURCES
ANALYTICS IN ACTION
CARTERTOONS
VIEW ALL +
QUICK POLL
VIEW ALL +