Remember this: Though originally designed for language learning, Smart.fm's software has been expanded; now the system can include user-generated material on topics such as the Internet memes shown above.
Smart.fm

Computing

An App so You'll Never Forget

Adaptive-learning algorithms calculate how often people need to see information to remember it.

  • Tuesday, November 3, 2009
  • By Erica Naone

A language-learning application that's already big in Japan is coming to the U.S. in the form of a new iPhone app. Smart.fm, based in Tokyo, says that the adaptive-learning algorithms behind its software can help users memorize all kinds of information.

Smart.fm is one of several companies selling software designed to help users remember. The company's algorithms were inspired by research that shows people remember information more effectively they try to memorize it at key times, says founder and chairman Andrew Smith Lewis.

Those algorithms determine how often to present a piece of information to the user and in what context. For example, a completely new word and its translation are shown frequently, and a user is asked relatively easy questions about them, designed to jog the memory. But once the user has demonstrated the ability to recall that word and its meaning, this information will appear less often.

"Efficiency is the main thing," Lewis says. "We want to optimize the sweet spot between the minimum number of times you have to see an item and the maximum effectiveness of that presentation."

Advertisement

To use Smart.fm, a person selects an existing list of material--a dictionary of foreign words, for instance--or starts building a new list. The list could be text-only, but the system also supports images and audio. A user might match the names of birds to the sounds they make, or view images of different parts of the human brain in order to learn how to identify them. Someone who snaps pictures of the people she meets at a conference might use the software to commit those people's names to memory.

"Learning applications which present stimuli adaptively based on the forgetting curve are not new but still relatively rare," says Peter Brusilovsky, director of the Personalized Adaptive Web Systems Lab at the University of Pittsburgh's School of Information Sciences. But Smart.fm's teaching methods can be applied to any type of material, Lewis says. "The adaptive-learning platform doesn't know if you're studying Russian painters or chess moves or French verbs," he says. "It just knows that these are individual objects."

Unlike other memory applications, Smart.fm takes a social approach, letting users share their lists and add comments to other lists. And in the future, Lewis says, there will be more ways to pull information into the system. The company is working on integrating with Freebase, a site that collects user-generated databases. Once the effort is complete, Smart.fm users who are interested in a particular topic should be able to access information about it from Freebase automatically.

"Education apps are one of the most interesting and growing areas of the iPhone app store," says Carl Howe, an analyst focusing on mobile research at the Yankee Group. Howe thinks Smart.fm was wise to broaden the scope of its material beyond just language learning. For education apps, he says, "the central aspect is knowing how to engage people's interest."

Howe notes that the top education apps for the iPhone are geared toward middle-school and elementary-school children. He believes there's a huge opportunity for college-level material, too. But companies designing e-learning apps may find themselves competing with material from established universities such as MIT and Stanford, which offer free material for self-directed study online.

Smart.fm's business model is based largely on the prospect of collaboration with other companies and institutions that want to offer online learning. The company has already partnered with the Japanese telecom giant NTT, which has used the software to create learning sites focusing on specific topics.

Lewis hopes that such deals will become Smart.fm's main source of revenue, though he also suggests that Smart.fm may offer premium content to users for a price. The iPhone app, however, is and will remain free.

Print

Related Articles

What's Next for the Netflix Algorithms?

Applying the lessons of the $1 million competition to other recommendation systems will be a challenge.

Adapting Websites to Users

Websites that change to fit the cognitive style of the user could be more effective at selling products online.

TR10: Modeling Surprise

Combining massive quantities of data, insights into human psychology, and machine learning can help manage surprising events, says Eric Horvitz.

Close Comments

To comment, please sign in or register

Forgot my password

GrantRobertson

3 Comments

  • 833 Days Ago
  • 11/03/2009

Literature review on Forgetting Curves

This is a subject of which I am keenly interested. I have written a literature review on Forgetting Curves, the fundamental neurobiological phenomenon underlying all of these memory enhancing programs. There are two different versions posted here:

http://www.ideationizing.com/2009/06/brief-history-of-mathematical.html

and here:

http://www.ideationizing.com/2009/06/brief-synopsis-of-long-history-of.html

I guess I should condense them into one article some day. Anyway, you can see that these ideas have been around since 1885.

Reply

rhutzul

8 Comments

  • 833 Days Ago
  • 11/03/2009

Curated Data...

Sounds like something the folks at Wolfram | Alpha should be looking at - a means by which to monetize the curated data driving their engine and a good way to offset the hurculean effort (and associated expense) required to gather and format said data...

Reply

drewmccormack

1 Comment

  • 833 Days Ago
  • 11/03/2009

Mental Case

It is worth pointing out that some apps that already exist on the iPhone are using these algorithms for learning.

I develop one of these apps: Mental Case. Mental Case is a flashcard application that has a built-in spaced repetition algorithm. It can be used for language learning, but also has support for images.

There is also a Mac version of Mental Case, which syncs with the iPhone app. The iPhone app also directly integrates with flashcardexchange.com for sharing content with other users.

Reply

korzonek

1 Comment

  • 814 Days Ago
  • 11/22/2009

omnimemo

I encourage you to check http://omnimemo.com. omnimemo was inspired by the research of Piotr Wozniak. BETA version of omnimemo was released almost two months ago and it combines results of research in the area of learning process optimization with a modern knowledge distribution platforms like web and in the near future mobile technology (iPhone, Android) ...and it is super easy to use!

Reply

Advertisement

MAGAZINE

Can We Build Tomorrow's Breakthroughs?

Manufacturing in the United States is in trouble. That's bad news not just for the country's economy but for the future of innovation.

Sponsored Content

Technologies from National Instruments

Adding Data Logging
Log measured data to a file and open it in Microsoft Excel

> Click here for more National Instruments Videos <
Whitepaper

Temperature Measurements with Thermocouples: How-To Guide

This document is part of the “How-To Guide for Most Common Measurements” centralized resource portal. This tutorial provides a detailed guide for measurement and device considerations to take temperature measurements using thermocouples. Get an introduction to thermocouples, which are inexpensive sensing devices widely used with PC-based data acquisition systems. Also review some specific thermocouple examples and learn how thermocouples work and ways to integrate them into a data acquisition measurement system.

View full PDF > Listen to story >
Find us on Youtube

Videos

A Robot Recruit that Can Do It All

More

Advertisement

Technology Review Lists

TR50

Our list of the 50 most innovative companies, including the following:

Lattice Power

Joule Unlimited

American Superconductor

Akamai

More

Advertisement

Facebook

Advertisement