Vidora Launches Novel Understandable Machine Learning Technology

Share Article

Vidora Cortex’s new understandable ML algorithms offer deep insight into Machine Learning models.

Vidora, the market leader in end-to-end Machine Learning (ML), announced today new Understandable ML technology which provides unprecedented insights into complex models. Understandable ML is an active area of research in the Machine Learning community, driven by an increasing need for businesses to unpack the internal logic of ML models. Vidora significantly advances that research with a set of innovative algorithms which solve substantial problems for businesses looking to deploy ML.

UNDERSTAND THE WHAT, WHY AND HOW OF MACHINE LEARNING MODELS

Many Machine Learning models function as black boxes. They take in data, output predictions, and provide little transparency as to what happens in between. According to a recent report by IBM, 60% of executives expressed concern over being able to explain the decision-making of AI systems, up from 29% in 2016.

In order to extract maximum value from their ML efforts, it’s critical that businesses have access not just to predictions, but also explanations. In particular, organizations need to understand what their model is predicting, why each prediction was made, and how that information can be used to pull the right business levers. Traditionally, ML has supplied only the “what”. Cortex, Vidora’s Self-Service ML platform, now provides insight into what, why, and how in an interface easily accessible to analysts, business intelligence specialists, and marketers.

This new technology is already being used by many of Vidora’s Fortune 500 partners. Example use cases include:

1) Commerce companies looking to understand which users are likely to buy an item (the what), which patterns of online user behavior indicated an intention to buy (the why), and how much a user’s likelihood to buy will increase if a particular behavior (e.g. time on site) is increased by 20% (the how).

2) Subscription media companies looking to understand who is likely to churn (the what), what types of content and usage patterns are most indicative of churn (the why), and how much predicted churn probability will decline if consumption of long-form drama is increased by 5% (the how).

3) Online real estate companies looking to understand who is most likely to buy a house in the next 30 days (the what), what online activity is indicative of their desire to buy (the why), and how much likelihood of buying will increase if a realtor reaches out (the how).

“Vidora’s Understandable ML technology has advanced how our team prioritizes strategic initiatives,” added Santosh Payal, Digital Analytics Reporting Manager at News Corp Australia. “Cortex goes a step further than other ML solutions by helping us understand the impact of various levers on our key business goals. Not only has Cortex given us a much better grasp on the health of our subscribers, it’s also sparked broader conversations on the best upstream strategies for retention.”

PUSHING THE BOUNDARIES OF ML RESEARCH

Vidora was founded by Ph.Ds in Machine Learning, and its technology continues to expand the frontiers of ML research. This latest technological breakthrough is one of several novel techniques that Vidora pioneered and built into its core ML platform.

“Vidora helps some of the largest global organizations solve their most pressing problems with ML”, said Abhik Majumdar, Vidora’s CTO who earned his Ph.D at University of California, Berkeley. “Our partners turn to us to develop sophisticated new technologies which drive tangible business value.”

LEARN MORE ABOUT CORTEX

By automating the hardest parts of the ML process and wrapping them in a simple interface, Cortex provides a powerful platform that anyone can use to build complex Machine Learning pipelines. With the latest release of Cortex, these pipelines become much more than automation tools, opening new doors for strategic initiatives across your business. Learn more about Vidora by filling out a contact form or emailing info@vidora.com.

ABOUT VIDORA

Vidora enables everyone to accomplish the incredible by making machine learning accessible to anyone in a business. Vidora makes machine learning simple, easy to use and intuitive by automating the most complex technical challenges of the machine learning pipeline. Our Self-Service Machine Learning platform, Cortex, is leveraged by some of the largest global brands like Walmart, News Corp, and Discovery.

Share article on social media or email:

View article via:

Pdf Print

Contact Author

Sarah Miller
info@vidora.com
+1 (415) 237-3423
Email >
@vidoracorp
since: 03/2012
Follow >
Vidora
Like >
Visit website