top of page
  • Writer's pictureJohn Ennis

Eye on AI - July 26th, 2019

Welcome to Aigora's "Eye on AI" series, where we round up exciting news at the intersection of consumer science and artificial intelligence!


This week we continue our previous discussion of the blurring line between online and offline retail, look at new progress on the consumer side of the Internet of Things (IoT), and delve more into potential applications of predictive modeling to food science.

Brick and Mortar Trends Toward E-Commerce

We begin with an article out of Daily Nation, titled “Retailers turn to e-commerce firms for sales boost”, where reporter Annie Njanja highlights a growing trend among local retailers and grocery outlets partnering with e-commerce firms to appeal to a rising number of young, tech-savvy shoppers.

“The trend was highlighted in the Nielsen Retail Measurement Defined FMCG Basket findings released last week,” continues Njanka, “which said that e-commerce would continue to evolve as the trade of fast-moving consumer goods (FMCG) online is projected to grow five times faster than offline transactions over the next five years.”

With three out of ten shoppers already willing to shop online, – a trend that will likely only grow over the next few years – brick and mortar companies merging with e-Commerce is beginning to seem more and more like a necessity, especially for those companies looking to expand their reach beyond the limitations of their physical address.

Low trust in e-commerce in certain locations could limit ROI early on, but as e-commerce expands and consumers become more familiar with the physical + e-commerce model, brick and mortar companies that don’t soon capitalize on the e-commerce trend may soon find themselves left in the dust by the competition.

Couple this news with our July 12 Eye on AI post for a more circumspect look on how Tech AI could revive brick and mortar retail.

Advances in the Internet of Things

Next, it’s news out of AI&IoT Daily on a new Smart Diaper System launched by Pampers, which is causing waves in the IoT news. The system, which Pampers is calling The Lumi, is a connected care system that combines a video monitor with an activity sensor placed on the baby’s diaper.

“The system, developed with Logitech and Verily, tracks sleep patterns, sleep routines, feeding and diapering, providing the information via mobile app,” writes reporter Chuck Martin. “The activity sensor tracks wet diapers and identifies in real time when a diaper needs to be changed.

This is intriguing tech, one with many other potential applications in the health and medical industries. For a deep dive into this subject in relation to the smart diaper, check out this Washington Post report.

And for more on how the smart diaper works, check out this video from Bernard Marr:

ML Breakthrough w/ Big Implications for Material Discoveries

Lastly, we turn to news from EurekAlert! about a promising new application of machine learning (ML) for innovative material discovery. The Toyko Institute of Technology released an article titled “Successful application of machine learning in the discovery of new polymers” that describes how a joint research group of scientists used an ML method involving ‘transfer learning’ to discover materials with desired properties, even from small datasets. In the case of the researchers, the data set they used was of polymeric properties, and the desired properties involved polymeric heat transfer.

“ML models on proxy properties were pre-trained where sufficient data were available on the related tasks,” the article reads, “these pre-trained models captured common features relevant to the target task… Applying this technique enabled the identification of thousands of promising ‘virtual’ polymers.”

The thermal conductivity of the new polymers was 80% higher than typical polyimides, a group of polyimides mass-produced for cookware, fuel cells, and other high-head applications since the 1950s. The application of this type of ‘transfer learning’ could potentially be a breakthrough in how scientists use machine learning to discover new desired materials in large or small data sets, adding to the advancement of products, food additives, medications and countless other applications.

“Through our project,” writes writes Junko Morikawa, Professor at Tokyo Institute of Technology’s School of Materials and Chemical Technology, “we aim to pursue not only the development of materials informatics but also contribute to fundamental advancement of materials science."

More on machine learning for material discovery on

Other news:


That's it for now. If you'd like to receive email updates from Aigora, including weekly video recaps of our blog activity, click on the button below to join our email list. Thanks for stopping by!

bottom of page