Interview John Myles White , Machine Learning for Hackers

Here is an interview with one of the younger researchers  and rock stars of the R Project, John Myles White,  co-author of Machine Learning for Hackers.

Ajay- What inspired you guys to write Machine Learning for Hackers. What has been the public response to the book. Are you planning to write a second edition or a next book?

John-We decided to write Machine Learning for Hackers because there were so many people interested in learning more about Machine Learning who found the standard textbooks a little difficult to understand, either because they lacked the mathematical background expected of readers or because it wasn’t clear how to translate the mathematical definitions in those books into usable programs. Most Machine Learning books are written for audiences who will not only be using Machine Learning techniques in their applied work, but also actively inventing new Machine Learning algorithms. The amount of information needed to do both can be daunting, because, as one friend pointed out, it’s similar to insisting that everyone learn how to build a compiler before they can start to program. For most people, it’s better to let them try out programming and get a taste for it before you teach them about the nuts and bolts of compiler design. If they like programming, they can delve into the details later.

We once said that Machine Learning for Hackers  is supposed to be a chemistry set for Machine Learning and I still think that’s the right description: it’s meant to get readers excited about Machine Learning and hopefully expose them to enough ideas and tools that they can start to explore on their own more effectively. It’s like a warmup for standard academic books like Bishop’s.
The public response to the book has been phenomenal. It’s been amazing to see how many people have bought the book and how many people have told us they found it helpful. Even friends with substantial expertise in statistics have said they’ve found a few nuggets of new information in the book, especially regarding text analysis and social network analysis — topics that Drew and I spend a lot of time thinking about, but are not thoroughly covered in standard statistics and Machine Learning  undergraduate curricula.
I hope we write a second edition. It was our first book and we learned a ton about how to write at length from the experience. I’m about to announce later this week that I’m writing a second book, which will be a very short eBook for O’Reilly. Stay tuned for details.

Ajay-  What are the key things that a potential reader can learn from this book?

John- We cover most of the nuts and bolts of introductory statistics in our book: summary statistics, regression and classification using linear and logistic regression, PCA and k-Nearest Neighbors. We also cover topics that are less well known, but are as important: density plots vs. histograms, regularization, cross-validation, MDS, social network analysis and SVM’s. I hope a reader walks away from the book having a feel for what different basic algorithms do and why they work for some problems and not others. I also hope we do just a little to shift a future generation of modeling culture towards regularization and cross-validation.

Ajay- Describe your journey as a science student up till your Phd. What are you current research interests and what initiatives have you done with them?

John-As an undergraduate I studied math and neuroscience. I then took some time off and came back to do a Ph.D. in psychology, focusing on mathematical modeling of both the brain and behavior. There’s a rich tradition of machine learning and statistics in psychology, so I got increasingly interested in ML methods during my years as a grad student. I’m about to finish my Ph.D. this year. My research interests all fall under one heading: decision theory. I want to understand both how people make decisions (which is what psychology teaches us) and how they should make decisions (which is what statistics and ML teach us). My thesis is focused on how people make decisions when there are both short-term and long-term consequences to be considered. For non-psychologists, the classic example is probably the explore-exploit dilemma. I’ve been working to import more of the main ideas from stats and ML into psychology for modeling how real people handle that trade-off. For psychologists, the classic example is the Marshmallow experiment. Most of my research work has focused on the latter: what makes us patient and how can we measure patience?

Ajay- How can academia and private sector solve the shortage of trained data scientists (assuming there is one)?

John- There’s definitely a shortage of trained data scientists: most companies are finding it difficult to hire someone with the real chops needed to do useful work with Big Data. The skill set required to be useful at a company like Facebook or Twitter is much more advanced than many people realize, so I think it will be some time until there are undergraduates coming out with the right stuff. But there’s huge demand, so I’m sure the market will clear sooner or later.

The changes that are required in academia to prepare students for this kind of work are pretty numerous, but the most obvious required change is that quantitative people need to be learning how to program properly, which is rare in academia, even in many CS departments. Writing one-off programs that no one will ever have to reuse and that only work on toy data sets doesn’t prepare you for working with huge amounts of messy data that exhibit shifting patterns. If you need to learn how to program seriously before you can do useful work, you’re not very valuable to companies who need employees that can hit the ground running. The companies that have done best in building up data teams, like LinkedIn, have learned to train people as they come in since the proper training isn’t typically available outside those companies.
Of course, on the flipside, the people who do know how to program well need to start learning more about theory and need to start to have a better grasp of basic mathematical models like linear and logistic regressions. Lots of CS students seem not to enjoy their theory classes, but theory really does prepare you for thinking about what you can learn from data. You may not use automata theory if you work at Foursquare, but you will need to be able to reason carefully and analytically. Doing math is just like lifting weights: if you’re not good at it right now, you just need to dig in and get yourself in shape.
About-
John Myles White is a Phd Student in  Ph.D. student in the Princeton Psychology Department, where he studies human decision-making both theoretically and experimentally. Along with the political scientist Drew Conway, he is  the author of a book published by O’Reilly Media entitled “Machine Learning for Hackers”, which is meant to introduce experienced programmers to the machine learning toolkit. He is also working with Mark Hansenon a book for laypeople about exploratory data analysis.John is the lead maintainer for several R packages, including ProjectTemplate and log4r.

(TIL he has played in several rock bands!)

—–
You can read more in his own words at his blog at http://www.johnmyleswhite.com/about/
He can be contacted via social media at Google Plus at https://plus.google.com/109658960610931658914 or twitter at twitter.com/johnmyleswhite/

Interview Dan Steinberg Founder Salford Systems

Here is an interview with Dan Steinberg, Founder and President of Salford Systems (http://www.salford-systems.com/ )

Ajay- Describe your journey from academia to technology entrepreneurship. What are the key milestones or turning points that you remember.

 Dan- When I was in graduate school studying econometrics at Harvard,  a number of distinguished professors at Harvard (and MIT) were actively involved in substantial real world activities.  Professors that I interacted with, or studied with, or whose software I used became involved in the creation of such companies as Sun Microsystems, Data Resources, Inc. or were heavily involved in business consulting through their own companies or other influential consultants.  Some not involved in private sector consulting took on substantial roles in government such as membership on the President’s Council of Economic Advisors. The atmosphere was one that encouraged free movement between academia and the private sector so the idea of forming a consulting and software company was quite natural and did not seem in any way inconsistent with being devoted to the advancement of science.

 Ajay- What are the latest products by Salford Systems? Any future product plans or modification to work on Big Data analytics, mobile computing and cloud computing.

 Dan- Our central set of data mining technologies are CART, MARS, TreeNet, RandomForests, and PRIM, and we have always maintained feature rich logistic regression and linear regression modules. In our latest release scheduled for January 2012 we will be including a new data mining approach to linear and logistic regression allowing for the rapid processing of massive numbers of predictors (e.g., one million columns), with powerful predictor selection and coefficient shrinkage. The new methods allow not only classic techniques such as ridge and lasso regression, but also sub-lasso model sizes. Clear tradeoff diagrams between model complexity (number of predictors) and predictive accuracy allow the modeler to select an ideal balance suitable for their requirements.

The new version of our data mining suite, Salford Predictive Modeler (SPM), also includes two important extensions to the boosted tree technology at the heart of TreeNet.  The first, Importance Sampled learning Ensembles (ISLE), is used for the compression of TreeNet tree ensembles. Starting with, say, a 1,000 tree ensemble, the ISLE compression might well reduce this down to 200 reweighted trees. Such compression will be valuable when models need to be executed in real time. The compression rate is always under the modeler’s control, meaning that if a deployed model may only contain, say, 30 trees, then the compression will deliver an optimal 30-tree weighted ensemble. Needless to say, compression of tree ensembles should be expected to be lossy and how much accuracy is lost when extreme compression is desired will vary from case to case. Prior to ISLE, practitioners have simply truncated the ensemble to the maximum allowable size.  The new methodology will substantially outperform truncation.

The second major advance is RULEFIT, a rule extraction engine that starts with a TreeNet model and decomposes it into the most interesting and predictive rules. RULEFIT is also a tree ensemble post-processor and offers the possibility of improving on the original TreeNet predictive performance. One can think of the rule extraction as an alternative way to explain and interpret an otherwise complex multi-tree model. The rules extracted are similar conceptually to the terminal nodes of a CART tree but the various rules will not refer to mutually exclusive regions of the data.

 Ajay- You have led teams that have won multiple data mining competitions. What are some of your favorite techniques or approaches to a data mining problem.

 Dan- We only enter competitions involving problems for which our technology is suitable, generally, classification and regression. In these areas, we are  partial to TreeNet because it is such a capable and robust learning machine. However, we always find great value in analyzing many aspects of a data set with CART, especially when we require a compact and easy to understand story about the data. CART is exceptionally well suited to the discovery of errors in data, often revealing errors created by the competition organizers themselves. More than once, our reports of data problems have been responsible for the competition organizer’s decision to issue a corrected version of the data and we have been the only group to discover the problem.

In general, tackling a data mining competition is no different than tackling any analytical challenge. You must start with a solid conceptual grasp of the problem and the actual objectives, and the nature and limitations of the data. Following that comes feature extraction, the selection of a modeling strategy (or strategies), and then extensive experimentation to learn what works best.

 Ajay- I know you have created your own software. But are there other software that you use or liked to use?

 Dan- For analytics we frequently test open source software to make sure that our tools will in fact deliver the superior performance we advertise. In general, if a problem clearly requires technology other than that offered by Salford, we advise clients to seek other consultants expert in that other technology.

 Ajay- Your software is installed at 3500 sites including 400 universities as per http://www.salford-systems.com/company/aboutus/index.html What is the key to managing and keeping so many customers happy?

 Dan- First, we have taken great pains to make our software reliable and we make every effort  to avoid problems related to bugs.  Our testing procedures are extensive and we have experts dedicated to stress-testing software . Second, our interface is designed to be natural, intuitive, and easy to use, so the challenges to the new user are minimized. Also, clear documentation, help files, and training videos round out how we allow the user to look after themselves. Should a client need to contact us we try to achieve 24-hour turn around on tech support issues and monitor all tech support activity to ensure timeliness, accuracy, and helpfulness of our responses. WebEx/GotoMeeting and other internet based contact permit real time interaction.

 Ajay- What do you do to relax and unwind?

 Dan- I am in the gym almost every day combining weight and cardio training. No matter how tired I am before the workout I always come out energized so locating a good gym during my extensive travels is a must. I am also actively learning Portuguese so I look to watch a Brazilian TV show or Portuguese dubbed movie when I have time; I almost never watch any form of video unless it is available in Portuguese.

 Biography-

http://www.salford-systems.com/blog/dan-steinberg.html

Dan Steinberg, President and Founder of Salford Systems, is a well-respected member of the statistics and econometrics communities. In 1992, he developed the first PC-based implementation of the original CART procedure, working in concert with Leo Breiman, Richard Olshen, Charles Stone and Jerome Friedman. In addition, he has provided consulting services on a number of biomedical and market research projects, which have sparked further innovations in the CART program and methodology.

Dr. Steinberg received his Ph.D. in Economics from Harvard University, and has given full day presentations on data mining for the American Marketing Association, the Direct Marketing Association and the American Statistical Association. After earning a PhD in Econometrics at Harvard Steinberg began his professional career as a Member of the Technical Staff at Bell Labs, Murray Hill, and then as Assistant Professor of Economics at the University of California, San Diego. A book he co-authored on Classification and Regression Trees was awarded the 1999 Nikkei Quality Control Literature Prize in Japan for excellence in statistical literature promoting the improvement of industrial quality control and management.

His consulting experience at Salford Systems has included complex modeling projects for major banks worldwide, including Citibank, Chase, American Express, Credit Suisse, and has included projects in Europe, Australia, New Zealand, Malaysia, Korea, Japan and Brazil. Steinberg led the teams that won first place awards in the KDDCup 2000, and the 2002 Duke/TeraData Churn modeling competition, and the teams that won awards in the PAKDD competitions of 2006 and 2007. He has published papers in economics, econometrics, computer science journals, and contributes actively to the ongoing research and development at Salford.

Interview Jaime Fitzgerald President Fitzgerald Analytics

Here is an interview with noted analytics expert Jaime Fitzgerald, of Fitzgerald Analytics.

Ajay-Describe your career journey from being a Harvard economist to being a text analytics thought leader.

 Jaime- I was attracted to economics because of the logic, the structured and systematic approach to understanding the world and to solving problems. In retrospect, this is the same passion for logic in problem solving that drives my business today.

About 15 years ago, I began working in consulting and initially took a traditional career path. I worked for well-known strategy consulting firms including First Manhattan Consulting Group, Novantas LLC, Braun Consulting, and for the former Japan-focused division of Deloitte Consulting, which had spun off as an independent entity. I was the only person in their New York City office for whom Japanese was not the first language.

While I enjoyed traditional consulting, I was especially passionate about the role of data, analytics, and process improvement. In traditional strategy consulting, these are important factors, but I had a vision for a “next generation” approach to strategy consulting that would be more transparent, more robust, and more focused on the role that information, analysis, and process plays in improving business results. I often explain that while my firm is “not your father’s consulting model,” we have incorporated key best practices from traditional consulting, and combined them with an approach that is more data-centric, technology-centric, and process-centric.

At the most fundamental level, I was compelled to found Fitzgerald Analytics more than six years ago by my passion for the role information plays in improving results, and ultimately improving lives. In my vision, data is an asset waiting to be transformed into results, including profit as well as other results that matter deeply to people. For example,one of the most fulfilling aspects of our work at Fitzgerald Analytics is our support of non-profits and social entrepreneurs, who we help increase their scale and their success in achieving their goals.

Ajay- How would you describe analytics as a career option to future students. What do you think are the most essential qualities an analytics career requires.

Jaime- My belief is that analytics will be a major driver of job-growth and career growth for decades. We are just beginning to unlock the full potential of analytics, and already the demand for analytic talent far exceeds the supply.

To succeed in analytics, the most important quality is logic. Many people believe that math or statistical skills are the most important quality, but in my experience, the most essential trait is what I call “ThoughtStyle” — critical thinking, logic, an ability to break down a problem into components, into sub-parts.

Ajay -What are your favorite techniques and methodologies in text analytics. How do you see social media and Big Data analytics as components of text analytics

 Jaime-We do a lot of work for our clients measuring Customer Experience, by which I mean the experience customers have when interacting with our clients. For example, we helped a major brokerage firm to measure 12 key “Moments that Matter,” including the operational aspects of customer service, customer satisfaction and sentiment, and ultimately customer behavior. Clients care about this a lot, because customer experience drives customer loyalty, which in turn drives customer behavior, customer loyalty, and customer profitability.

Text analytics plays a key role in these projects because much of our data on customer sentiment comes via unstructured text data. For example, we have access to call center transcripts and notes, to survey responses, and to social media comments.

We use a variety of methods, some of which I’m not in a position to describe in great detail. But at a high level, I would say that our favorite text analytics methodologies are “hybrid solutions” which use a two-step process to answer key questions for clients:

Step 1: convert unstructured data into key categorical variables (for example, using contextual analysis to flag users who are critical vs. neutral vs. advocates)

Step 2: linking sentiment categories to customer behavior and profitability (for example, linking customer advocacy and loyalty with customer profits as well as referral volume, to define the ROI that clients accrue for customer satisfaction improvements)

Ajay- Describe your consulting company- Fitzgerald Analytics and some of the work that you have been engaged in.

 Jaime- Our mission is to “illuminate reality” using data and to convert Data to Dollars for our clients. We have a track record of doing this well, with concrete and measurable results in the millions of dollars. As a result, 100% of our clients have engaged us for more than one project: a 100% client loyalty rate.

Our specialties–and most frequent projects–include customer profitability management projects, customer segmentation, customer experience management, balanced scorecards, and predictive analytics. We are often engaged to address high-stakes analytic questions, including issues that help to set long-term strategy. In other cases, clients hire us to help them build their internal capabilities. We have helped build several brand new analytic teams for clients, which continue to generate millions of dollars of profits with their fact-based recommendations.

Our methodology is based on Steven Covey’s principle: “begin with the end in mind,” the concept of starting with the client’s goal and working backwards from there. I often explain that our methods are what you would have gotten if Steven Covey had been a data analyst…we are applying his principles to the world of data analytics.

Ajay- Analytics requires more and more data while privacy requires the least possible data. What do you think are the guidelines that need to be built in sharing internet browsing and user activity data and do we need regulations just like we do for sharing financial data.

 Jaime- Great question. This is an essential challenge of the big data era. My perspective is that firms who depend on user data for their analysis need to take responsibility for protecting privacy by using data management best practices. Best practices to adequately “mask” or remove private data exist…the problem is that these best practices are often not applied. For example, Facebook’s practice of sharing unique user IDs with third-party application companies has generated a lot of criticism, and could have been avoided by applying data management best practices which are well known among the data management community.

If I were able to influence public policy, my recommendation would be to adopt a core set of simple but powerful data management standards that would protect consumers from perhaps 95% of the privacy risks they face today. The number one standard would be to prohibit sharing of static, personally identifiable user IDs between companies in a manner that creates “privacy risk.” Companies can track unique customers without using a static ID…they need to step up and do that.

Ajay- What are your favorite text analytics software that you like to work with.

 Jaime- Because much of our work in deeply embedded into client operations and systems, we often use the software our clients already prefer. We avoid recommending specific vendors unless our client requests it. In tandem with our clients and alliance partners, we have particular respect for Autonomy, Open Text, Clarabridge, and Attensity.

Biography-

http://www.fitzgerald-analytics.com/jaime_fitzgerald.html

The Founder and President of Fitzgerald Analytics, Jaime has developed a distinctively quantitative, fact-based, and transparent approach to solving high stakes problems and improving results.  His approach enables translation of Data to Dollars™ using methodologies clients can repeat again and again.  He is equally passionate about the “human side of the equation,” and is known for his ability to link the human and the quantitative, both of which are needed to achieve optimal results.

Experience: During more than 15 years serving clients as a management strategy consultant, Jaime has focused on customer experience and loyalty, customer profitability, technology strategy, information management, and business process improvement.  Jaime has advised market-leading banks, retailers, manufacturers, media companies, and non-profit organizations in the United States, Canada, and Singapore, combining strategic analysis with hands-on implementation of technology and operations enhancements.

Career History: Jaime began his career at First Manhattan Consulting Group, specialists in financial services, and was later a Co-Founder at Novantas, the strategy consultancy based in New York City.  Jaime was also a Manager for Braun Consulting, now part of Fair Isaac Corporation, and for Japan-based Abeam Consulting, now part of NEC.

Background: Jaime is a graduate of Harvard University with a B.A. in Economics.  He is passionate and supportive of innovative non-profit organizations, their effectiveness, and the benefits they bring to our society.

Upcoming Speaking Engagements:   Jaime is a frequent speaker on analytics, information management strategy, and data-driven profit improvement.  He recently gave keynote presentations on Analytics in Financial Services for The Data Warehousing Institute, the New York Technology Council, and the Oracle Financial Services Industry User Group. A list of Jaime’s most interesting presentations on analyticscan be found here.

He will be presenting a client case study this fall at Text Analytics World re:   “New Insights from ‘Big Legacy Data’: The Role of Text Analytics” 

Connecting with Jaime:  Jaime can be found at Linkedin,  and Twitter.  He edits the Fitzgerald Analytics Blog.

Harvard DropOut Writes Open Letter- His Startup has 350m users

Note from Mark “zucken”berg

An Open Letter from Facebook Founder Mark Zuckerberg
by Mark Zuckerberg Yesterday at 9:23pm

It has been a great year for making the world more open and connected. Thanks to your help, more than 350 million people around the world are using Facebook to share their lives online.

To make this possible, we have focused on giving you the tools you need to share and control your information. Starting with the very first version of Facebook five years ago, we’ve built tools that help you control what you share with which individuals and groups of people. Our work to improve privacy continues today.

Facebook’s current privacy model revolves around “networks” — communities for your school, your company or your region. This worked well when Facebook was mostly used by students, since it made sense that a student might want to share content with their fellow students.

Over time people also asked us to add networks for companies and regions as well. Today we even have networks for some entire countries, like India and China.

However, as Facebook has grown, some of these regional networks now have millions of members and we’ve concluded that this is no longer the best way for you to control your privacy. Almost 50 percent of all Facebook users are members of regional networks, so this is an important issue for us. If we can build a better system, then more than 100 million people will have even more control of their information.

The plan we’ve come up with is to remove regional networks completely and create a simpler model for privacy control where you can set content to be available to only your friends, friends of your friends, or everyone.

We’re adding something that many of you have asked for — the ability to control who sees each individual piece of content you create or upload. In addition, we’ll also be fulfilling a request made by many of you to make the privacy settings page simpler by combining some settings. If you want to read more about this, we began discussing this plan back in July.

Since this update will remove regional networks and create some new settings, in the next couple of weeks we’ll ask you to review and update your privacy settings. You’ll see a message that will explain the changes and take you to a page where you can update your settings. When you’re finished, we’ll show you a confirmation page so you can make sure you chose the right settings for you. As always, once you’re done you’ll still be able to change your settings whenever you want.

We’ve worked hard to build controls that we think will be better for you, but we also understand that everyone’s needs are different. We’ll suggest settings for you based on your current level of privacy, but the best way for you to find the right settings is to read through all your options and customize them for yourself. I encourage you to do this and consider who you’re sharing with online.

Thanks for being a part of making Facebook what it is today, and for helping to make the world more open and connected.

Well atleast he can write open code.

MS Smacks Google Docs with Slideshare

Our favorite drop outs from the Phd Program just learned that they should not moon the giant. The company founded in Paul Allen building at Stanford, also known as Gogol /Google announced they would create a Cloud OS with much fan fare. Only to find their own cloud prodocutivity offering Google Docs bested by Slideshare.

Now you can import your Gmail attachments Google docs into slideshare, for much better professional sharing within your office.

Here is an embedded SlideShare ppt called Google Hacks, note the much better visual appeal in this vis a vis your Google Docs.

Well as for the Stanford dropouts this is what happens when you dont complete your Phd education.

Citation- http://www.slideshare.net/rickdog/google-hacks

As per Cloud Computing and Office productivity goes,

Harvard Dropouts (Microsoft) 1- Stanford Dropouts ( Google) 0

Unless Google creates a cloud version of Open Office- but who needs that anyway?

who needs search- just ctrl F

Google Hacks

View more documents from rickdog.
Disclaimer- The author uses Google Docs extensively. If you are from Google. Please do not block his Gmail id , guys.
Academic Disclaimer-The author intends to complete his Phd. these are his personal views only.