�ݺ�ߣshows by User: mrogati

�ݺ�ߣshows by User: mrogati / http://www.slideshare.net/images/logo.gif �ݺ�ߣshows by User: mrogati / Mon, 12 Mar 2012 15:58:23 GMT �ݺ�ߣShare feed for �ݺ�ߣshows by User: mrogati The Model and the Train Wreck - A Training Data How-To -- @mrogati's talk at Strata 2012 /slideshow/the-model-and-the-train-wreck-a-training-data-howto-mrogatis-talk-at-strata-2012/11978330 modeltrainwreck-120312155825-phpapp02
Getting training data for a recommender system is easy: if users clicked it, it’s a positive – if they didn’t, it’s a negative. … Or is it? You’ve probably learned an algorithm to run on top of your existing algorithm, now and every time you re-train. And what do you do when the data product you’re building doesn’t have any users yet? Do you really launch with random results, hand label 50K examples, or ask a Turker to pretend they’re User #1337? Unlike having a better algorithm, having better training data can improve your results by orders of magnitude. Yet training data generation is often an afterthought—a footnote in a formula-filled publication. In this talk, we use examples from production recommender systems to bring training data to the forefront: from overcoming presentation bias to the art of crowdsourcing subjective judgments to creative data exhaust exploitation and feature creation.]]>
Getting training data for a recommender system is easy: if users clicked it, it’s a positive – if they didn’t, it’s a negative. … Or is it? You’ve probably learned an algorithm to run on top of your existing algorithm, now and every time you re-train. And what do you do when the data product you’re building doesn’t have any users yet? Do you really launch with random results, hand label 50K examples, or ask a Turker to pretend they’re User #1337? Unlike having a better algorithm, having better training data can improve your results by orders of magnitude. Yet training data generation is often an afterthought—a footnote in a formula-filled publication. In this talk, we use examples from production recommender systems to bring training data to the forefront: from overcoming presentation bias to the art of crowdsourcing subjective judgments to creative data exhaust exploitation and feature creation.]]> Mon, 12 Mar 2012 15:58:23 GMT /slideshow/the-model-and-the-train-wreck-a-training-data-howto-mrogatis-talk-at-strata-2012/11978330 mrogati@slideshare.net(mrogati) The Model and the Train Wreck - A Training Data How-To -- @mrogati's talk at Strata 2012 mrogati Getting training data for a recommender system is easy: if users clicked it, it’s a positive – if they didn’t, it’s a negative. … Or is it? You’ve probably learned an algorithm to run on top of your existing algorithm, now and every time you re-train. And what do you do when the data product you’re building doesn’t have any users yet? Do you really launch with random results, hand label 50K examples, or ask a Turker to pretend they’re User #1337? Unlike having a better algorithm, having better training data can improve your results by orders of magnitude. Yet training data generation is often an afterthought—a footnote in a formula-filled publication. In this talk, we use examples from production recommender systems to bring training data to the forefront: from overcoming presentation bias to the art of crowdsourcing subjective judgments to creative data exhaust exploitation and feature creation. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/modeltrainwreck-120312155825-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds" /> Getting training data for a recommender system is easy: if users clicked it, it’s a positive – if they didn’t, it’s a negative. … Or is it? You’ve probably learned an algorithm to run on top of your existing algorithm, now and every time you re-train. And what do you do when the data product you’re building doesn’t have any users yet? Do you really launch with random results, hand label 50K examples, or ask a Turker to pretend they’re User #1337? Unlike having a better algorithm, having better training data can improve your results by orders of magnitude. Yet training data generation is often an afterthought—a footnote in a formula-filled publication. In this talk, we use examples from production recommender systems to bring training data to the forefront: from overcoming presentation bias to the art of crowdsourcing subjective judgments to creative data exhaust exploitation and feature creation.

The Model and the Train Wreck - A Training Data How-To -- @mrogati's talk at Strata 2012 from Monica Rogati

]]> 7402 10 https://cdn.slidesharecdn.com/ss_thumbnails/modeltrainwreck-120312155825-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds presentation White http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

1 Feeding The Vizard: finding the stories in the data /mrogati/feeding-the-vizard-finding-the-stories-in-the-data feedingthevizard-111104161752-phpapp02
Visualization helps us tell data stories - but before that happens, how do you find these stories?]]>
Visualization helps us tell data stories - but before that happens, how do you find these stories?]]> Fri, 04 Nov 2011 16:17:51 GMT /mrogati/feeding-the-vizard-finding-the-stories-in-the-data mrogati@slideshare.net(mrogati) Feeding The Vizard: finding the stories in the data mrogati Visualization helps us tell data stories - but before that happens, how do you find these stories? <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/feedingthevizard-111104161752-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds" /> Visualization helps us tell data stories - but before that happens, how do you find these stories?

Feeding The Vizard: finding the stories in the data from Monica Rogati

]]> 1196 7 https://cdn.slidesharecdn.com/ss_thumbnails/feedingthevizard-111104161752-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds presentation White http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

1 Lies, damned lies and the data scientist 2011 strata summit /slideshow/lies-damned-lies-and-the-data-scientist-2011-strata-summit-9343968/9343968 liesdamnedliesandthedatascientist-2011stratasummit-110920101210-phpapp01
When it comes to big data insights, how do you know you’re asking the right questions? Hiring data scientists is a good start – we’re seeing their growth both on LinkedIn and at LinkedIn. But even data scientists are not immune from the myriad of hidden pitfalls that keep your key insights out of sight. Drawing from a deceptively simple exercise that I’ve used to haze dozens of data scientists on their first day, I will discuss the good, the bad and the ugly lessons we’ve learned about asking the right questions, denominators and being a data skeptic.]]>
When it comes to big data insights, how do you know you’re asking the right questions? Hiring data scientists is a good start – we’re seeing their growth both on LinkedIn and at LinkedIn. But even data scientists are not immune from the myriad of hidden pitfalls that keep your key insights out of sight. Drawing from a deceptively simple exercise that I’ve used to haze dozens of data scientists on their first day, I will discuss the good, the bad and the ugly lessons we’ve learned about asking the right questions, denominators and being a data skeptic.]]> Tue, 20 Sep 2011 10:12:07 GMT /slideshow/lies-damned-lies-and-the-data-scientist-2011-strata-summit-9343968/9343968 mrogati@slideshare.net(mrogati) Lies, damned lies and the data scientist 2011 strata summit mrogati When it comes to big data insights, how do you know you’re asking the right questions? Hiring data scientists is a good start – we’re seeing their growth both on LinkedIn and at LinkedIn. But even data scientists are not immune from the myriad of hidden pitfalls that keep your key insights out of sight. Drawing from a deceptively simple exercise that I’ve used to haze dozens of data scientists on their first day, I will discuss the good, the bad and the ugly lessons we’ve learned about asking the right questions, denominators and being a data skeptic. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/liesdamnedliesandthedatascientist-2011stratasummit-110920101210-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds" /> When it comes to big data insights, how do you know you’re asking the right questions? Hiring data scientists is a good start – we’re seeing their growth both on LinkedIn and at LinkedIn. But even data scientists are not immune from the myriad of hidden pitfalls that keep your key insights out of sight. Drawing from a deceptively simple exercise that I’ve used to haze dozens of data scientists on their first day, I will discuss the good, the bad and the ugly lessons we’ve learned about asking the right questions, denominators and being a data skeptic.

Lies, damned lies and the data scientist 2011 strata summit from Monica Rogati

]]> 1852 2 https://cdn.slidesharecdn.com/ss_thumbnails/liesdamnedliesandthedatascientist-2011stratasummit-110920101210-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

1 Sequencing the Startup DNA - @mrogati's talk at StartupFest 2011 /slideshow/sequencing-the-startup-dna-mrogatis-talk/9090012 sequencingthestartupdna-110831201632-phpapp02
What makes entrepreneurs different, and where do they come from? Are they born or taught? Are they unusually mobile in their careers? Does geography play a role? Do mentors and relationships matter? Numerous studies explore these questions by surveying hundreds of entrepreneurs. At LinkedIn, we take a different approach, on a different scale. By sifting through more than 120 million public profiles, we can analyze tens of thousands startup founder# profiles - and find common threads linking their careers. (make sure you follow along w/ the speaker notes; my slides are minimalist.)]]>
What makes entrepreneurs different, and where do they come from? Are they born or taught? Are they unusually mobile in their careers? Does geography play a role? Do mentors and relationships matter? Numerous studies explore these questions by surveying hundreds of entrepreneurs. At LinkedIn, we take a different approach, on a different scale. By sifting through more than 120 million public profiles, we can analyze tens of thousands startup founder# profiles - and find common threads linking their careers. (make sure you follow along w/ the speaker notes; my slides are minimalist.)]]> Wed, 31 Aug 2011 20:16:28 GMT /slideshow/sequencing-the-startup-dna-mrogatis-talk/9090012 mrogati@slideshare.net(mrogati) Sequencing the Startup DNA - @mrogati's talk at StartupFest 2011 mrogati What makes entrepreneurs different, and where do they come from? Are they born or taught? Are they unusually mobile in their careers? Does geography play a role? Do mentors and relationships matter? Numerous studies explore these questions by surveying hundreds of entrepreneurs. At LinkedIn, we take a different approach, on a different scale. By sifting through more than 120 million public profiles, we can analyze tens of thousands startup founder# profiles - and find common threads linking their careers. (make sure you follow along w/ the speaker notes; my slides are minimalist.) <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/sequencingthestartupdna-110831201632-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds" /> What makes entrepreneurs different, and where do they come from? Are they born or taught? Are they unusually mobile in their careers? Does geography play a role? Do mentors and relationships matter? Numerous studies explore these questions by surveying hundreds of entrepreneurs. At LinkedIn, we take a different approach, on a different scale. By sifting through more than 120 million public profiles, we can analyze tens of thousands startup founder# profiles - and find common threads linking their careers. (make sure you follow along w/ the speaker notes; my slides are minimalist.)

Sequencing the Startup DNA - @mrogati's talk at StartupFest 2011 from Monica Rogati

]]> 11320 9 https://cdn.slidesharecdn.com/ss_thumbnails/sequencingthestartupdna-110831201632-phpapp02-thumbnail.jpg?width=120&height=120&fit=bounds presentation White http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

1 Tiger Moms, Ninjas and Chips - Oh My! Uncovering the story in the data - @mrogati's talk at Ignite Google I/O 2011 /slideshow/ignite-google-io-2011/7985960 monicarogatiignitegoogleio2011-110516153047-phpapp01
These are the (slightly edited) slides for my 2011 Ignite Google I/O talk. See comments for video. About Ignite: ignite.oreilly.com ]]>
These are the (slightly edited) slides for my 2011 Ignite Google I/O talk. See comments for video. About Ignite: ignite.oreilly.com ]]> Mon, 16 May 2011 15:30:42 GMT /slideshow/ignite-google-io-2011/7985960 mrogati@slideshare.net(mrogati) Tiger Moms, Ninjas and Chips - Oh My! Uncovering the story in the data - @mrogati's talk at Ignite Google I/O 2011 mrogati These are the (slightly edited) slides for my 2011 Ignite Google I/O talk. See comments for video. About Ignite: ignite.oreilly.com <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/monicarogatiignitegoogleio2011-110516153047-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds" /> These are the (slightly edited) slides for my 2011 Ignite Google I/O talk. See comments for video. About Ignite: ignite.oreilly.com

Tiger Moms, Ninjas and Chips - Oh My! Uncovering the story in the data - @mrogati's talk at Ignite Google I/O 2011 from Monica Rogati

]]> 1016 5 https://cdn.slidesharecdn.com/ss_thumbnails/monicarogatiignitegoogleio2011-110516153047-phpapp01-thumbnail.jpg?width=120&height=120&fit=bounds presentation White http://activitystrea.ms/schema/1.0/post

http://activitystrea.ms/schema/1.0/posted

1 https://cdn.slidesharecdn.com/profile-photo-mrogati-48x48.jpg?cb=1523082842 . www.rogati.com https://cdn.slidesharecdn.com/ss_thumbnails/modeltrainwreck-120312155825-phpapp02-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/the-model-and-the-train-wreck-a-training-data-howto-mrogatis-talk-at-strata-2012/11978330 The Model and the Trai... https://cdn.slidesharecdn.com/ss_thumbnails/feedingthevizard-111104161752-phpapp02-thumbnail.jpg?width=320&height=320&fit=bounds mrogati/feeding-the-vizard-finding-the-stories-in-the-data Feeding The Vizard: fi... https://cdn.slidesharecdn.com/ss_thumbnails/liesdamnedliesandthedatascientist-2011stratasummit-110920101210-phpapp01-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/lies-damned-lies-and-the-data-scientist-2011-strata-summit-9343968/9343968 Lies, damned lies and ...