In the past, our analytics tools were proprietary, product-oriented solutions. These were necessarily limited in flexibility and they locked customers into the slow innovation cycles and whims of vendors. These closed-source solutions forced a “one size fits all” approach to analytics with monolithic tools that did not offer easy customization for different needs.

Open Data Science has changed that. It offers innovative software—free of proprietary restrictions and tailorable for all varieties of data science teams—created in the transparent collaboration that is driving today’s tech boom.

Science fiction author Arthur C. Clarke once wrote, “any sufficiently advanced technology is indistinguishable from magic.”

We’re nearer than ever to that incomprehensible, magical future. Our gadgets understand our speech, driverless cars have made their debut and we’ll soon be viewing virtual worlds at home.

These “magical” technologies spring from a 21st-century spirit of innovation—but not only from big companies. Thanks to the Internet—and to the open source movement—companies of all sizes are able to spur advancements in science and technology.

It’s no different for advanced analytics. And it’s about time.

In the past, our analytics tools were proprietary, product-oriented solutions. These were necessarily limited in flexibility and they locked customers into the slow innovation cycles and whims of vendors. These closed-source solutions forced a “one size fits all” approach to analytics with monolithic tools that did not offer easy customization for different needs.

Open Data Science has changed that. It offers innovative software—free of proprietary restrictions and tailorable for all varieties of data science teams—created in the transparent collaboration that is driving today’s tech boom.

The Magic 8-Ball of Automated Modeling

One of Open Data Science’s most visible points of innovation is in the sphere of data science modeling.

Initially, models were created exclusively by statisticians and analysts for business professionals, but demand from the business sector for software that could do this job gave rise to automatic model fitting—often called “black box” analytics—in which analysts let software algorithmically generate models that fit data and create predictive models.

Such a system creates models, but much like a magic 8-ball, it offers its users answers without business explanations. Mysteries are fun for toys, but no business will bet on them. Quite understandably, no marketing manager or product manager wants to approach the CEO with predictions, only to be stumped when he asks how the manager arrived at them. As Clarke knew, it’s not really magic creating the models, it’s advanced technology and it too operates under assumptions that might or might not make sense for the business.

App Starters Means More Transparent Modeling

Today’s business professionals want faster time-to-value and are dazzled by advanced technologies like automated model fitting, but they also want to understand exactly how and why the work.

That’s why Continuum Analytics is hard at work on Open Data Science solutions including Anaconda App Starters, expected to debut later this year. App Starters are solution “templates” aimed to be a 60-80 percent data science solution that make it easy for businesses to have a starting point. App Starters serve the same purpose as the “black box”—faster time-to-value— but are not a “black box” in that it allows analysts to see exactly how the model was created and to tweak models as desired.

Because the App Starters are are based on Open Data Science, they don’t include proprietary restrictions that keep business professionals or data scientists in the dark regarding the analytics pipeline including the algorithms. It still provides the value of “automagically” creating models, but the details of how it does so are transparent and accessible to the team. With App Starters, business professionals will finally have confidence in the models they’re using to formulate business strategies, while getting faster time-to-value from their growing data.

Over time App Starters will get more sophisticated and will include recommendations—just like how Netflix offers up movie and tv show recommendations for your watching pleasure—that will learn and suggest algorithms and visualizations that best fit the data. Unlike “black boxes” the entire narrative as to why recommendations are offered will be available for the business analyst to learn and gain confidence in the recommendations. However, the business analyst can choose to use the recommendation, tweak the recommendation, use the template without recommendations or they could try tuning the suggested models to find a perfect fit. This type of innovation will further the advancement of sophisticated data science solutions that realize more business value, while instilling confidence in the solution.  

Casting Spells with Anaconda

Although App Starters are about to shake up automated modeling, businesses require melding new ideas with tried-and-true solutions. In business analytics, for instance, tools like Microsoft Excel are a staple of the field and being able to integrate them with newer “magic” is highly desirable.

Fortunately, interoperability is one of the keystones of the Open Data Science philosophy and Anaconda provides a way to bridge the reliable old world with the magical new one. With Anaconda, analysts who are comfortable using Excel have an entry point into the world of predictive analytics from the comfort of their spreadsheets. By using the same familiar interface, analysts can access powerful Python libraries to apply cutting-edge analytics to their data. Anaconda recognizes that business analysts want to improve—not disrupt—a proven workflow.

Because Anaconda leverages the Python ecosystem, analysts using Anaconda will achieve powerful results. They might apply a formula to an Excel sheet with a million data rows to predict repeat customers or they may create beautiful, informative visualizations to show how sales have shifted to a new demographic after the company’s newest marketing campaign kicked off. With Anaconda, business analysts can continue using Excel as their main interface, while harnessing the newest “magic” available in the open source community.

Open Data Science for Wizards…and Apprentices

Open Data Science is an inclusive movement. Although open source languages like Python and R dominate data science and allow for the most advanced—and therefore “magical”—analytics technology available, the community is open to all levels of expertise.

Anaconda is a great way for business analysts, for example, to embark on the road toward advanced analytics. But solutions, like App Starters, give advanced wizards the algorithmic visibility to alter and improve models as they see fit.

Open Data Science gives us the “sufficiently advanced technology” that Arthur C. Clarke mentioned—but it puts the power of that magic in our hands.


About the Author

We help people discover, analyze, and collaborate by connecting their curiosity and experience with any data. It all comes together here. Anaconda gives superpowers to people who change the world.

Read more

Join the Disucssion