【MIT新型交互式机器学习预测工具让人人都能拥有AI”超能力“】来自MIT和布朗大学的研究人员,为他们的“北极星”交互式数据系统增加了一项新功能,使该系统可以“立即生成机器学习模型”,只要和任何可以访问的现有数据库一起使用,就能生成有用的预测。研究人员将此功能称为 “虚拟数据科学家”,它可以让人们在相关技术资源有限的环境中也能享受到机器学习技术的便利,例如让普通医生、大多数中小型企业、私营咖啡店和零售商等显然不拥有专门数据专家团队也拥有数据洞察力。


MIT’s new interactive machine learning prediction tool could give everyone AI superpowers

Soon, you might not need anything more specialized than a readily accessible touchscreen device and any existing data sets you have access to in order to build powerful prediction tools. A new experiment from MIT and Brown Universityresearchers have added a capability to their ‘Northstar’ interactive data system that can “instantly generate machine-learning models” to use with their exiting data sets in order to generate useful predictions.

One example the researchers provide is that doctors could make use of the system to make predictions about the likelihood their patients have of contracting specific diseases based on their medial history. Or, they suggest, a business owner could use their historical sales data to develop more accurate forecasts, quickly and without a ton of manual analytics work.

Researchers are calling this feature the Northstar system’s “virtual data scientist,” (or VDS) and it sounds like it could actually replace the human equivalent, especially in settings where one would never actually be readily available or resourced anyway. Your average doctor’s office doesn’t have a dedicated data scientist headcount, for instance, and nor do most small- to medium-sized businesses for that matter. Independently owned and operated coffee shops and retailers definitely wouldn’t otherwise have access to this kind of insight.

This new tool is built on automated machine-learning techniques that are becoming much more ‘au courant,’ since it helps expand the number of people for whom AI technology is accessible.

Northstar itself is the product of more than four years of work, and presents a blank canvas that’s compatible across multiple platforms, and then users can upload their own data sets, which show up as boxes on the interface. They can then drag and drop those into the centre area of the canvas and then draw connecting lines to indicate to that they should be processed with an algorithm of their choosing in combination with one another.

So basically, they could theoretically grab a dataset detailing metabolic rates of patients, and another one detailing their age, and then derive from that how often a specific disease occurs across those two factors. Now, with the new virtual data scientist feature, they’ll be able to combine inputs to generate predictive, AI-based analysis across these combined factors as well.

Researchers have also designed this VDS system so that it’s actually the fastest application of automated machine learning to date. That’s another key piece for making it usable by everyone, since it’s not really feasible to imagine people working with this digital whitetable and then waiting ours for results to come out. Next up, it’s going to improve error reporting to help ensure that non-specialist users not only find it easy to use, but also get clear indicators when they do something wrong so they can fix it next time.


Comments are closed.