Prescribing mHealth Apps, AI for Data Science & a Trading Platform for Sensor Data (#WIC)

By December 8, 2014 Week in Context No Comments

This week in context, we highlight the prescription of mHealth applications in the UK, hear about the ‘Automated Statistician’ and pizza menu that reads your eyes and suggest some wearable maps for the holidays.

This Week in Context

Your Weekly Update on All Things Context, November 28 2014

Quoted

“Mobile phones are now ubiquitous in developing countries, with 89 active subscriptions per 100 inhabitants. Though many types of population data are scarce in developing countries, the metadata generated by millions of mobile phones and recorded by mobile phone operators can enable unprecedented insights about individuals and societies. Used with appropriate restraint, this data has great potential for good, including immediate use in the fight against Ebola.”

Enabling Humanitarian Use of Mobile Phone Data
Yves-Alexandre de Montjoye, Jake Kendall and Cameron F. Kerry

Five Must-Reads

1. Take Two Apps for That and Call Me in the Morning

UK doctors are being encouraged to prescribe mHealth apps to help tackle depression, especially amongst teens. Treatment-by-app is a way of meeting them on familiar ground. The digital initiative will also involve lifelogging and health-focused smart devices, and the National Health Service is giving those apps that offer cognitive behavioural therapy priority as it comes to screening for official approval.

2. Automatic Statistician: AI for Data Science

Google is funding Automatic Statistician, a project that bills itself as an artificial intelligence for datascience and aims to automate the selection, building and explanation of machine learning models.

In a nutshell, Automatic Statistician works by looking at a dataset and then determining which type of model would be best for analyzing it as well as which features, or variables, are the strongest. After the model runs, Automatic Statistician will return a text report explaining its findings in plain English — or as close as you can get when dealing with statistics.

3. A trading platform for sensor data

Terbine wants to become a data broker for the world of connected devices. The Las Vegas based startup set out to build a platform where companies can buy, sell and share the data their sensors are collecting. More on Gigaom.com, as Terbine’s website is still in ‘stealth’ mode. (Hat tip: Yves Blondeel)

4. Which pizza do you desire? The ‘Subconscious Menu’

Screen Shot 2014-12-08 at 11.37.34Pizza Hut’s ‘Subconscious Menu’ uses eye-tracking technology to see which of 20 different ingredients that you are looking at you like the most. It then takes all of three seconds to identify the pizza you want based on which you looked at the longest. Pizza Hut says its Subconscious Menu is still in trials, but after testing to a 98 percent success rate Engadget reports, it may eventually appear in restaurants. (Hat tip: Tanguy De Lestré)

5. Wearable Maps 

If you are still looking for that perfect gift for a geo-data-affectionado, or just someone who loves original fashion and their home city, this might just be what you need: custom clothes printed with maps by Monochrome.

Monochrome allows you to customise tank tops, pencil skirts, flare skirts and tshirts using OpenStreet Map data.

On the right, you see what an Antwerp-styled tank top would look like. Pretty neat!

A work of Art vs Tech: Louvre Museum DNA

The Louvre Museum’s DNA is a study of visitors’ behaviour by the MIT Senseable Lab that measured visiting sequences and duration using Bluetooth data for 24,452 unique mobile devices: “The sensors recorded a unique encrypted identifier that distinguishes each Bluetooth-enabled mobile device within its range, as well as time stamps for entry and exit times. Assuming that a mobile device belongs to a person, we can relate the movement of the device to that of the visitor.” Watch the simulation video here.

Interesting to note is that 8.2% of visitors had bluetooth turned on. (Bluetooth-enabled devices measured vs ticket sales.)

louvre4_louvre_museum_dna_bluetooth_tracking_data

Image courtesy MIT Senseable City Lab and the Louvre Museum.

1. Entrance
2. Psyche & Cupid
3. Captive by Michelangelo
4. Gallery Daru
5. Aphrodite ‘Venus de Milo’
6. The winged Victory of Samothrace
7. Big gallery
8. Mona Lisa
9. The Wounded Cuirassier

Blue: bluetooth sensors
Yellow: shorter stay visitors (less than 1.5 hours)
Orange: longer stay visitors (more than 6 hours)

Papers, Talks & Research

  • An analysis of visitors’ behavior in The Louvre Museum: a study using Bluetooth data (design, planning, sensors, crowd analysis, paper)
  • Visual Sentiment Prediction with Deep Convolutional Neural Networks (machine learning, neural network, sentiment, emotion, paper)
  • Factoring Variations in Natural Images with Deep Gaussian Mixture Models (deep learning, machine learning, image recognition, paper)
  • Enabling Humanitarian Use of Mobile Phone Data (mobile, privacy, big data, metadata, paper)

Have a great and productive week!
(and if you haven’t done so yet, kindly consider subscribing to the Week in Context here.)

Ann

Leave a Reply

Contact us
×